results: 本文的方法可以在 meta-learning、personalized federated learning、end-to-end learning 和 Wasserstein distributionally robust optimization with side information（WDRO-SI）等应用中实现高效的优化。特别是在 Stochastic Nonconvex Optimization 中，我们的方法与现有的下界匹配。在数学实验中，我们的方法的复杂性不依赖于任务数量。Abstract
We introduce contextual stochastic bilevel optimization (CSBO) -- a stochastic bilevel optimization framework with the lower-level problem minimizing an expectation conditioned on some contextual information and the upper-level decision variable. This framework extends classical stochastic bilevel optimization when the lower-level decision maker responds optimally not only to the decision of the upper-level decision maker but also to some side information and when there are multiple or even infinite many followers. It captures important applications such as meta-learning, personalized federated learning, end-to-end learning, and Wasserstein distributionally robust optimization with side information (WDRO-SI). Due to the presence of contextual information, existing single-loop methods for classical stochastic bilevel optimization are unable to converge. To overcome this challenge, we introduce an efficient double-loop gradient method based on the Multilevel Monte-Carlo (MLMC) technique and establish its sample and computational complexities. When specialized to stochastic nonconvex optimization, our method matches existing lower bounds. For meta-learning, the complexity of our method does not depend on the number of tasks. Numerical experiments further validate our theoretical results.
摘要
我们介绍了 Contextual Stochastic Bilevel Optimization（CSBO）框架，这是一种带有上下文信息的随机二重优化框架，下一级决策者根据上一级决策者的决策以及一些副信息进行优化。这种框架超越了经典的随机二重优化，因为下一级决策者不仅响应上一级决策者的决策，还响应一些副信息，并且可能有多个或无数多个追随者。它涵盖了重要的应用，如meta-学习、个性化联合学习、端到端学习以及 Wasserstein Distributionally Robust Optimization with Side Information（WDRO-SI）。由于上下文信息的存在，传统的单循环方法无法收敛。为解决这个挑战，我们提出了一种高效的双循环梯度法，基于 Multilevel Monte-Carlo（MLMC）技术，并证明其样本和计算复杂度。当特化到随机非 convex 优化时，我们的方法与已有的下界匹配。对于 meta-学习，我们的方法的复杂度不виси于任务数量。实验数据也 validate 我们的理论结果。
Feature Selection in the Contrastive Analysis Setting
results: 实验结果表明，CFS方法在四个数据集中均可以比前所有的超参数和无监督特征选择方法表现更好，并且可以更好地捕捉到CA设置中的特征差异。Abstract
Contrastive analysis (CA) refers to the exploration of variations uniquely enriched in a target dataset as compared to a corresponding background dataset generated from sources of variation that are irrelevant to a given task. For example, a biomedical data analyst may wish to find a small set of genes to use as a proxy for variations in genomic data only present among patients with a given disease (target) as opposed to healthy control subjects (background). However, as of yet the problem of feature selection in the CA setting has received little attention from the machine learning community. In this work we present contrastive feature selection (CFS), a method for performing feature selection in the CA setting. We motivate our approach with a novel information-theoretic analysis of representation learning in the CA setting, and we empirically validate CFS on a semi-synthetic dataset and four real-world biomedical datasets. We find that our method consistently outperforms previously proposed state-of-the-art supervised and fully unsupervised feature selection methods not designed for the CA setting. An open-source implementation of our method is available at https://github.com/suinleelab/CFS.
摘要
contrastive analysis (CA) 指的是对target数据集中具有特点的变化进行探索，相比较background数据集中的无关变化。例如，生物医学数据分析员可能想要找到一小组基因作为病人群体疾病（target）中唯一存在的变化表示，而不是健康控制Subject（background）。然而，当前的CA设定中的特征选择问题尚未得到机器学习社区的足够关注。在这篇文章中，我们提出了对CA设定进行特征选择的方法——对比特征选择（CFS）。我们通过对CA设定的信息理论分析来证明我们的方法，并在一个半 sintetic数据集和四个实际生物医学数据集上进行了验证。我们发现，我们的方法常常超越了已有的supervised和完全无监督特征选择方法，不是CA设定。我们的实现可以在https://github.com/suinleelab/CFS上获取。
Learning to design protein-protein interactions with enhanced generalization
paper_authors: Anton Bushuiev, Roman Bushuiev, Anatolii Filkin, Petr Kouba, Marketa Gabrielova, Michal Gabriel, Jiri Sedlar, Tomas Pluskal, Jiri Damborsky, Stanislav Mazurenko, Josef Sivic
for: 提高生物医学研究和开发新药的进步，发现加强蛋白质-蛋白质交互（PPI）的突变是关键。
methods: 使用机器学习方法，特别是SE(3)-等变征模型，以实现大规模学习和泛化。
results: 提出了PPIRef数据集，这是世界上最大、非重复的蛋白质-蛋白质交互数据集，并使用PPIRef数据集预训练PPIformer模型，并通过调整预训练损失函数来预测蛋白质-蛋白质交互突变的效果。最终，通过比较其他现有的状态之最好方法，提高了新的PPIformer方法的泛化性。Abstract
Discovering mutations enhancing protein-protein interactions (PPIs) is critical for advancing biomedical research and developing improved therapeutics. While machine learning approaches have substantially advanced the field, they often struggle to generalize beyond training data in practical scenarios. The contributions of this work are three-fold. First, we construct PPIRef, the largest and non-redundant dataset of 3D protein-protein interactions, enabling effective large-scale learning. Second, we leverage the PPIRef dataset to pre-train PPIformer, a new SE(3)-equivariant model generalizing across diverse protein-binder variants. We fine-tune PPIformer to predict effects of mutations on protein-protein interactions via a thermodynamically motivated adjustment of the pre-training loss function. Finally, we demonstrate the enhanced generalization of our new PPIformer approach by outperforming other state-of-the-art methods on new, non-leaking splits of standard labeled PPI mutational data and independent case studies optimizing a human antibody against SARS-CoV-2 and increasing the thrombolytic activity of staphylokinase.
摘要
发现加强蛋白蛋白交互（PPI）的突变可以推动生物医学研究和开发改进的药物。虽然机器学习方法在这个领域已经做出了很大的进步，但它们经常在实际应用场景中难以泛化。这个研究的贡献有三个方面：1. 我们构建了3D蛋白蛋白交互的最大和非重复的数据集PPIRef，使得大规模学习成为可能。2. 我们利用PPIRef数据集来预训练PPIformer，一种新的SE(3)-可变型模型，可以在不同的蛋白蛋白绑定变体中进行广泛的泛化。3. 我们利用PPIformer模型来预测蛋白蛋白交互中突变的影响，通过在预训练损失函数中做一种热力学激活的调整。最后，我们表明了我们的新PPIformer方法在新的非泄漏分组的标注的PPI突变数据集和独立的案例研究中表现出了更高的泛化能力，比如人类抗体 against SARS-CoV-2和提高破坏酶的溶解活性。
Preventing Language Models From Hiding Their Reasoning
results: 研究发现，当 LLM 强度增加时，它们更容易使用编码推理来解决问题，但这些推理步骤可能不可读明白于人类读者。此外，研究还提出了一种方法来评估防御机制，并证明在某些条件下，重叠 rewrite 可以成功防止模型编码推理。Abstract
Large language models (LLMs) often benefit from intermediate steps of reasoning to generate answers to complex problems. When these intermediate steps of reasoning are used to monitor the activity of the model, it is essential that this explicit reasoning is faithful, i.e. that it reflects what the model is actually reasoning about. In this work, we focus on one potential way intermediate steps of reasoning could be unfaithful: encoded reasoning, where an LLM could encode intermediate steps of reasoning in the generated text in a way that is not understandable to human readers. We show that language models can be trained to make use of encoded reasoning to get higher performance without the user understanding the intermediate steps of reasoning. We argue that, as language models get stronger, this behavior becomes more likely to appear naturally. Finally, we describe a methodology that enables the evaluation of defenses against encoded reasoning, and show that, under the right conditions, paraphrasing successfully prevents even the best encoding schemes we built from encoding more than 3 bits of information per KB of text.
摘要
大型语言模型（LLM）经常从中间步骤的理解来生成复杂问题的答案。当这些中间步骤的理解用于监测模型的活动时，则必须确保这些Explicit reasoning是 faithful，即模型真正正在理解什么。在这项工作中，我们关注一种可能的不忠的中间步骤理解情况：编码理解，其中一个LLM可能将中间步骤的理解编码到生成的文本中，以至于人类读者无法理解。我们证明了语言模型可以通过编码理解来提高性能，而无需用户理解中间步骤的理解。我们还 argues that，随着语言模型的强大化，这种行为越来越容易出现。最后，我们描述了一种方法ология，用于评估防御机制 против编码理解，并显示，在合适的条件下，重叠成功地防止了我们构建的最佳编码方案中更多于3 bits的信息。
Multi-fidelity Design of Porous Microstructures for Thermofluidic Applications
paper_authors: Jonathan Tammer Eweis-LaBolle, Chuanning Zhao, Yoonjin Won, Ramin Bostanabad
for: 这 paper 的目的是为了设计最佳的热管理解决方案，以满足现代电子设备的高效热管理需求。
methods: 这 paper 使用了数据驱动的方法，利用特征函数（SDFs）来编码设计空间，并通过多元模拟和优化算法来找到最佳的热管理解决方案。
results: 这 paper 的结果显示，使用这种数据驱动的方法可以快速和有效地找到最佳的热管理解决方案，并且可以满足现代电子设备的高效热管理需求。Abstract
As modern electronic devices are increasingly miniaturized and integrated, their performance relies more heavily on effective thermal management. Two-phase cooling methods enhanced by porous surfaces, which capitalize on thin-film evaporation atop structured porous surfaces, are emerging as potential solutions. In such porous structures, the optimum heat dissipation capacity relies on two competing objectives that depend on mass and heat transfer. The computational costs of evaluating these objectives, the high dimensionality of the design space which a voxelated microstructure representation, and the manufacturability constraints hinder the optimization process for thermal management. We address these challenges by developing a data-driven framework for designing optimal porous microstructures for cooling applications. In our framework we leverage spectral density functions (SDFs) to encode the design space via a handful of interpretable variables and, in turn, efficiently search it. We develop physics-based formulas to quantify the thermofluidic properties and feasibility of candidate designs via offline simulations. To decrease the reliance on expensive simulations, we generate multi-fidelity data and build emulators to find Pareto-optimal designs. We apply our approach to a canonical problem on evaporator wick design and obtain fin-like topologies in the optimal microstructures which are also characteristics often observed in industrial applications.
摘要
现代电子设备逐渐减小和集成，其性能受到有效热管理的依赖程度加大。两相冷却方法，通过使用结构化孔隙表面增强薄膜蒸发，被视为可能的解决方案。在such porous structures中，最佳热耗抑制 capacitance rely on two competing objectives that depend on mass and heat transfer。计算这些目标的成本，高维度的设计空间的computational cost，以及制造性能约束，使得优化过程受到阻碍。我们解决这些挑战 by developing a data-driven framework for designing optimal porous microstructures for cooling applications。在我们的框架中，我们利用spectral density functions (SDFs)来编码设计空间，通过一些可读性好的变量来快速搜索。我们开发了物理学基于的方程来评估候选设计的热流体性能和可行性，并通过offline simulations来验证。为了减少成本expensive simulations，我们生成多 fideltysimulation data和建立模型来找到Pareto优化的设计。我们在一个标准的蒸发器柱状结构设计问题中应用我们的方法，并获得了fin-like topologies的优化结构，这些结构也经常出现在工业应用中。
Understanding and Improving Ensemble Adversarial Defense
methods: 该论文使用了一种新的错误理论来解释 ensemble adversarial defense 的效果，并提出了一种名为 interactive global adversarial training (iGAT) 的新方法来提高 ensemble adversarial defense 的性能。
results: 根据实验结果，iGAT 可以在 CIFAR10 和 CIFAR100 datasets 下提高 ensemble adversarial defense 的性能，最高提升达 17%，并在 white-box 和 black-box 攻击下都有显著的提升。Abstract
The strategy of ensemble has become popular in adversarial defense, which trains multiple base classifiers to defend against adversarial attacks in a cooperative manner. Despite the empirical success, theoretical explanations on why an ensemble of adversarially trained classifiers is more robust than single ones remain unclear. To fill in this gap, we develop a new error theory dedicated to understanding ensemble adversarial defense, demonstrating a provable 0-1 loss reduction on challenging sample sets in an adversarial defense scenario. Guided by this theory, we propose an effective approach to improve ensemble adversarial defense, named interactive global adversarial training (iGAT). The proposal includes (1) a probabilistic distributing rule that selectively allocates to different base classifiers adversarial examples that are globally challenging to the ensemble, and (2) a regularization term to rescue the severest weaknesses of the base classifiers. Being tested over various existing ensemble adversarial defense techniques, iGAT is capable of boosting their performance by increases up to 17% evaluated using CIFAR10 and CIFAR100 datasets under both white-box and black-box attacks.
摘要
《 ensemble 策略在对抗攻击方面变得受欢迎，它在多个基础分类器之间进行协作来防止对抗攻击。 DESPITE 这种策略的实际成功，对 ensemble 对抗防御的理论解释仍然不清楚。 To fill this gap, we develop a new error theory dedicated to understanding ensemble adversarial defense, demonstrating a provable 0-1 loss reduction on challenging sample sets in an adversarial defense scenario. Guided by this theory, we propose an effective approach to improve ensemble adversarial defense, named interactive global adversarial training (iGAT). The proposal includes (1) a probabilistic distributing rule that selectively allocates adversarial examples to different base classifiers that are globally challenging to the ensemble, and (2) a regularization term to rescue the severest weaknesses of the base classifiers. tested over various existing ensemble adversarial defense techniques, iGAT is capable of boosting their performance by increases up to 17% evaluated using CIFAR10 and CIFAR100 datasets under both white-box and black-box attacks.》Note: Please note that the translation is in Simplified Chinese, and the formatting of the text may be different from the original English version.
Parameter-Efficient Methods for Metastases Detection from Clinical Notes
results: 我们的最佳模型在 F1 分数上达到 73.8%，准确率为 84%，并 recall 为 65.8%。Abstract
Understanding the progression of cancer is crucial for defining treatments for patients. The objective of this study is to automate the detection of metastatic liver disease from free-style computed tomography (CT) radiology reports. Our research demonstrates that transferring knowledge using three approaches can improve model performance. First, we utilize generic language models (LMs), pretrained in a self-supervised manner. Second, we use a semi-supervised approach to train our model by automatically annotating a large unlabeled dataset; this approach substantially enhances the model's performance. Finally, we transfer knowledge from related tasks by designing a multi-task transfer learning methodology. We leverage the recent advancement of parameter-efficient LM adaptation strategies to improve performance and training efficiency. Our dataset consists of CT reports collected at Memorial Sloan Kettering Cancer Center (MSKCC) over the course of 12 years. 2,641 reports were manually annotated by domain experts; among them, 841 reports have been annotated for the presence of liver metastases. Our best model achieved an F1-score of 73.8%, a precision of 84%, and a recall of 65.8%.
摘要
理解肿瘤的进程对于定义患者的治疗是非常重要。本研究的目标是自动从自由式 computed tomography（CT） radiology report中检测到肝肿瘤病变。我们的研究表明，通过三种方法可以提高模型的性能。首先，我们使用通用语言模型（LM），先前在自我超vised的方式进行预训练。第二，我们使用自动注释大量未标注数据集来培训我们的模型，这种方法显著提高了模型的性能。最后，我们利用相关任务的知识传递来设计多任务转移学习方法。我们利用了最近的参数效率LM参照扩展策略，以提高性能和训练效率。我们的数据集包括 Memorial Sloan Kettering Cancer Center（MSKCC）在12年期间收集的CT报告2,641份，其中841份已经被培训出版物的培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培训出版物培�
Minimax Optimal Submodular Optimization with Bandit Feedback
results: 本文获得了一个新的下界 bound，该 bound 表示在 $T$ 次循环中， learner 的 regret 将是 $\mathcal{O}(\min_{i \le k}(in^{1/3}T^{2/3} + \sqrt{n^{k-i}T}))$。此外，本文还提出了一个可以对下界 bound 做出匹配的算法。Abstract
We consider maximizing a monotonic, submodular set function $f: 2^{[n]} \rightarrow [0,1]$ under stochastic bandit feedback. Specifically, $f$ is unknown to the learner but at each time $t=1,\dots,T$ the learner chooses a set $S_t \subset [n]$ with $|S_t| \leq k$ and receives reward $f(S_t) + \eta_t$ where $\eta_t$ is mean-zero sub-Gaussian noise. The objective is to minimize the learner's regret over $T$ times with respect to ($1-e^{-1}$)-approximation of maximum $f(S_*)$ with $|S_*| = k$, obtained through greedy maximization of $f$. To date, the best regret bound in the literature scales as $k n^{1/3} T^{2/3}$. And by trivially treating every set as a unique arm one deduces that $\sqrt{ {n \choose k} T }$ is also achievable. In this work, we establish the first minimax lower bound for this setting that scales like $\mathcal{O}(\min_{i \le k}(in^{1/3}T^{2/3} + \sqrt{n^{k-i}T}))$. Moreover, we propose an algorithm that is capable of matching the lower bound regret.
摘要
我们考虑最大化一个单调、下调的集合函数 $f：2^{[n]} \to [0,1]$ 在数据随机弹指示下。具体来说，$f$ 是学习者不知道的，但在每个时间 $t=1,\dots,T$ 中，学习者选择一个集合 $S_t \subset [n]$ ， $|S_t| \leq k$，并获得奖励 $f(S_t) + \eta_t$，其中 $\eta_t$ 是mean-zero sub-Gaussian 噪声。学习者的目标是在 $T$ 次时间内，与 ($1-e^{-1}$) 近似最大的 $f(S_*)$ ，其中 $|S_*| = k$，通过简单的单调最大化 $f$ 而得到。现有的最差 regret bound 是 $k n^{1/3} T^{2/3}$。而通过将每个集合视为单一的枪一样，则可以得到 $\sqrt{n \choose k} T}$ 的 regret bound。在这个研究中，我们建立了首个最小最差下界，其scale如 $\mathcal{O}(\min_{i \le k}(in^{1/3}T^{2/3} + \sqrt{n^{k-i}T}))$。此外，我们也提出了一个能够匹配下界的算法。
Approximate Heavy Tails in Offline (Multi-Pass) Stochastic Gradient Descent
results: 研究发现，随着数据点数量的增加，offline SGD 会变得越来越 “power-law-like”，而 online SGD 则会保持不变。此外，研究还在 synthetic data 和神经网络上进行了实验来证明理论结论。Abstract
A recent line of empirical studies has demonstrated that SGD might exhibit a heavy-tailed behavior in practical settings, and the heaviness of the tails might correlate with the overall performance. In this paper, we investigate the emergence of such heavy tails. Previous works on this problem only considered, up to our knowledge, online (also called single-pass) SGD, in which the emergence of heavy tails in theoretical findings is contingent upon access to an infinite amount of data. Hence, the underlying mechanism generating the reported heavy-tailed behavior in practical settings, where the amount of training data is finite, is still not well-understood. Our contribution aims to fill this gap. In particular, we show that the stationary distribution of offline (also called multi-pass) SGD exhibits 'approximate' power-law tails and the approximation error is controlled by how fast the empirical distribution of the training data converges to the true underlying data distribution in the Wasserstein metric. Our main takeaway is that, as the number of data points increases, offline SGD will behave increasingly 'power-law-like'. To achieve this result, we first prove nonasymptotic Wasserstein convergence bounds for offline SGD to online SGD as the number of data points increases, which can be interesting on their own. Finally, we illustrate our theory on various experiments conducted on synthetic data and neural networks.
摘要
现在的一些实验研究表明，SGD可能在实际情况下具有重尾性行为，并且这种重尾性与总性能之间存在相关性。在这篇论文中，我们探究了这种重尾性的出现。先前的研究只考虑了在线（也称为单通道）SGD，其中证明重尾性的存在需要训练数据的无穷多余。因此，在实际设置中，其下面机制仍然不很清楚。我们的贡献是填充这个空白。具体来说，我们表明了离线（也称为多通道）SGD的站点分布 exhibits 'approximate' 的力学尾部，并且这种预测错误是通过如何让Empirical distribution of training data converge to the true underlying data distribution在 Wasserstein 度量下控制的。我们的主要结论是，随着数据点数量增加，离线 SGD 会逐渐具有 'power-law-like' 的行为。为了实现这个结论，我们首先证明了离线 SGD 与在线 SGD 的非尺寸 Wasserstein 准确性 bound，这可能是一个独立的兴趣点。最后，我们在synthetic data和神经网络上进行了多个实验来证明我们的理论。
Bayesian Optimization with Hidden Constraints via Latent Decision Models
results: 我们通过数值实验表示，HC-LSBO在 synthetic 和实际数据集上表现出色，特别是在大规模警区设置问题上。与基线方法相比，HC-LSBO 提供了显著的性能和效率改善。Abstract
Bayesian optimization (BO) has emerged as a potent tool for addressing intricate decision-making challenges, especially in public policy domains such as police districting. However, its broader application in public policymaking is hindered by the complexity of defining feasible regions and the high-dimensionality of decisions. This paper introduces the Hidden-Constrained Latent Space Bayesian Optimization (HC-LSBO), a novel BO method integrated with a latent decision model. This approach leverages a variational autoencoder to learn the distribution of feasible decisions, enabling a two-way mapping between the original decision space and a lower-dimensional latent space. By doing so, HC-LSBO captures the nuances of hidden constraints inherent in public policymaking, allowing for optimization in the latent space while evaluating objectives in the original space. We validate our method through numerical experiments on both synthetic and real data sets, with a specific focus on large-scale police districting problems in Atlanta, Georgia. Our results reveal that HC-LSBO offers notable improvements in performance and efficiency compared to the baselines.
摘要
bayesian 优化（BO）已成为复杂决策挑战的强大工具，尤其在公共政策领域such as 警区划分。然而，它在公共政策决策中的更广泛应用受到定义可行区域的复杂性和决策的高维度所阻碍。这篇文章介绍了隐藏的约束 latent Space Bayesian 优化（HC-LSBO），一种 integrate Bayesian 优化方法和秘密决策模型。这种方法利用一种变换自动编码器来学习原始决策空间中的可行分布，从而实现原始空间和隐藏空间之间的两个方向的映射。由此，HC-LSBO 捕捉了公共政策中隐藏的约束，允许在隐藏空间进行优化而不影响原始空间中的目标评价。我们通过对 sintetic 和实际数据集进行数学实验，发现HC-LSBO 对基eline 提供了显著的性能和效率提升。
M3C: A Framework towards Convergent, Flexible, and Unsupervised Learning of Mixture Graph Matching and Clustering
paper_authors: Jiaxin Lu, Zetian Jiang, Tianzhe Wang, Junchi Yan
for: 这篇论文targets real-world graph matching and clustering tasks, where graphs exhibit diverse modes and require grouping before or along with matching.
methods: 该方法基于Minorize-Maximization框架，提供了学习自由的 guarantee of theoretical convergence, along with relaxed clustering for enhanced flexibility.
results: 实验结果表明，该方法在公共benchmark上的准确率和效率都高于现状的graph matching和mixture graph matching和分 clustering方法。Here’s the English version for reference:
for: This paper targets real-world graph matching and clustering tasks, where graphs exhibit diverse modes and require grouping before or along with matching.
methods: The method is based on the Minorize-Maximization framework, providing learning-free guarantees of theoretical convergence, along with relaxed clustering for enhanced flexibility.
results: Experimental results demonstrate that our method outperforms state-of-the-art graph matching and mixture graph matching and clustering approaches in both accuracy and efficiency on public benchmarks.Abstract
Existing graph matching methods typically assume that there are similar structures between graphs and they are matchable. However, these assumptions do not align with real-world applications. This work addresses a more realistic scenario where graphs exhibit diverse modes, requiring graph grouping before or along with matching, a task termed mixture graph matching and clustering. We introduce Minorize-Maximization Matching and Clustering (M3C), a learning-free algorithm that guarantees theoretical convergence through the Minorize-Maximization framework and offers enhanced flexibility via relaxed clustering. Building on M3C, we develop UM3C, an unsupervised model that incorporates novel edge-wise affinity learning and pseudo label selection. Extensive experimental results on public benchmarks demonstrate that our method outperforms state-of-the-art graph matching and mixture graph matching and clustering approaches in both accuracy and efficiency. Source code will be made publicly available.
摘要
现有的图匹配方法通常假设图有相似结构，可以匹配。然而，这些假设与实际应用场景不符。本工作面临现实世界中图表现多种模式的问题，需要在匹配之前或同时进行图分组，一种被称为杂合图匹配和分群。我们介绍了一种不含学习的算法，名为小于最大化匹配和分群（M3C），该算法 garantías了理论上的收敛，并提供了放宽分群的灵活性。基于M3C，我们开发了一种无监督的模型，名为UM3C，它包括新的边绑定学习和 Pseudo标签选择。经过广泛的实验研究，我们发现OUR方法在公共测试 benchmark上比现有的图匹配和杂合图匹配和分群方法更高效和更准确。代码将在公共平台上发布。
Bridging Distributionally Robust Learning and Offline RL: An Approach to Mitigate Distribution Shift and Partial Data Coverage
results: 我们的提议算法在实验中表现出优于现有算法，并且我们通过对单个策略强度的假设进行Characterization sample complexity。Abstract
The goal of an offline reinforcement learning (RL) algorithm is to learn optimal polices using historical (offline) data, without access to the environment for online exploration. One of the main challenges in offline RL is the distribution shift which refers to the difference between the state-action visitation distribution of the data generating policy and the learning policy. Many recent works have used the idea of pessimism for developing offline RL algorithms and characterizing their sample complexity under a relatively weak assumption of single policy concentrability. Different from the offline RL literature, the area of distributionally robust learning (DRL) offers a principled framework that uses a minimax formulation to tackle model mismatch between training and testing environments. In this work, we aim to bridge these two areas by showing that the DRL approach can be used to tackle the distributional shift problem in offline RL. In particular, we propose two offline RL algorithms using the DRL framework, for the tabular and linear function approximation settings, and characterize their sample complexity under the single policy concentrability assumption. We also demonstrate the superior performance our proposed algorithm through simulation experiments.
摘要
文本：offline reinforcement learning（RL）算法的目标是使用历史数据学习优化策略，不能直接访问环境进行在线探索。offline RL中的一个主要挑战是分布转移，即数据生成策略对应的状态动作访问分布与学习策略之间的差异。许多最近的研究使用了偏见的想法开发了offline RL算法，并对其样本复杂性进行了定量化。与offline RL文献不同，分布robust学习（DRL）领域提供了一个理性的框架，使用最小最大形式来处理训练和测试环境之间的模型差异。本文想要将这两个领域相连，我们表明了DRL框架可以解决offline RL中的分布转移问题。特别是，我们提出了两种使用DRL框架的offline RL算法，一种是 для tabular设置，另一种是 для线性函数近似设置，并对其样本复杂性进行了定量化。我们还通过实验证明了我们的提出算法的优秀性。Translation:文本：The goal of an offline reinforcement learning (RL) algorithm is to learn optimal policies using historical (offline) data, without access to the environment for online exploration. One of the main challenges in offline RL is the distribution shift, which refers to the difference between the state-action visitation distribution of the data generating policy and the learning policy. Many recent works have used the idea of pessimism for developing offline RL algorithms and characterizing their sample complexity under a relatively weak assumption of single policy concentrability. Different from the offline RL literature, the area of distributionally robust learning (DRL) offers a principled framework that uses a minimax formulation to tackle model mismatch between training and testing environments. In this work, we aim to bridge these two areas by showing that the DRL approach can be used to tackle the distributional shift problem in offline RL. In particular, we propose two offline RL algorithms using the DRL framework, for the tabular and linear function approximation settings, and characterize their sample complexity under the single policy concentrability assumption. We also demonstrate the superior performance of our proposed algorithm through simulation experiments.
methods: 这个论文提出了一种新的方法 called Minority Class Rebalancing through Augmentation by Generative modeling (MCRAGE)，它利用一种深度生成模型来生成高质量的人工数据样本，以增强原有的异质数据集，从而实现更好的数据均衡。
results: 作者们通过对MCRAGE方法与其他方法进行比较，发现MCRAGE方法可以提高医疗机器学习模型的准确率和F1分数，并且可以减少对不同性别和年龄等敏感属性的偏见。Abstract
In the field of healthcare, electronic health records (EHR) serve as crucial training data for developing machine learning models for diagnosis, treatment, and the management of healthcare resources. However, medical datasets are often imbalanced in terms of sensitive attributes such as race/ethnicity, gender, and age. Machine learning models trained on class-imbalanced EHR datasets perform significantly worse in deployment for individuals of the minority classes compared to samples from majority classes, which may lead to inequitable healthcare outcomes for minority groups. To address this challenge, we propose Minority Class Rebalancing through Augmentation by Generative modeling (MCRAGE), a novel approach to augment imbalanced datasets using samples generated by a deep generative model. The MCRAGE process involves training a Conditional Denoising Diffusion Probabilistic Model (CDDPM) capable of generating high-quality synthetic EHR samples from underrepresented classes. We use this synthetic data to augment the existing imbalanced dataset, thereby achieving a more balanced distribution across all classes, which can be used to train an unbiased machine learning model. We measure the performance of MCRAGE versus alternative approaches using Accuracy, F1 score and AUROC. We provide theoretical justification for our method in terms of recent convergence results for DDPMs with minimal assumptions.
摘要
在医疗领域，电子医疗记录（EHR）作为重要的训练数据，用于发展 диагности学、治疗和医疗资源管理的机器学习模型。然而，医疗数据经常受到敏感属性的影响，如种族/民族、性别和年龄，这会导致机器学习模型在部署过程中对少数群体的性能下降，从而可能导致不公正的医疗结果。为解决这个挑战，我们提出了少数类重新平衡通过扩充（MCRAGE），一种使用深度生成模型生成高质量的人工数据来增强不平衡的数据集的新方法。MCRAGE过程中，我们首先训练一个 Conditional Denoising Diffusion Probabilistic Model（CDDPM），可以生成来自少数类的高质量人工数据。然后，我们使用这些人工数据来扩充现有的不平衡数据集，以达到更平衡的分布，可以用于训练不偏袋机器学习模型。我们使用精度、F1分数和AUROC三个指标来衡量MCRAGE的性能和相对于其他方法的比较。此外，我们还提供了对MCRAGE方法的理论 justify，基于最近的DDPM的征untuous结果和最小的假设。
results: 本文证明了这些定义稳定性之间存在等价关系，并证明了一些扩充稳定性的结果，以便更好地理解和掌握最近几年出现的多种稳定性概念。Abstract
We show that many definitions of stability found in the learning theory literature are equivalent to one another. We distinguish between two families of definitions of stability: distribution-dependent and distribution-independent Bayesian stability. Within each family, we establish equivalences between various definitions, encompassing approximate differential privacy, pure differential privacy, replicability, global stability, perfect generalization, TV stability, mutual information stability, KL-divergence stability, and R\'enyi-divergence stability. Along the way, we prove boosting results that enable the amplification of the stability of a learning rule. This work is a step towards a more systematic taxonomy of stability notions in learning theory, which can promote clarity and an improved understanding of an array of stability concepts that have emerged in recent years.
摘要
我们证明了学习理论中多种稳定性定义之间存在等价关系。我们将稳定性定义分为两类：受分布影响的和不受分布影响的极 bayesian稳定性。每个家族中，我们证明了各种定义之间的等价关系，包括近似隐私、纯隐私、复制性、全局稳定性、完美泛化、TV稳定性、信息稳定性和KL散度稳定性。在过程中，我们证明了增强结果，使得稳定性的学习规则得到扩大。这项工作是一步 towards 更系统化的学习理论中稳定性概念的分类，可以促进清晰和学习许多年来出现的稳定性概念的更好的理解。
Fast Machine Learning Method with Vector Embedding on Orthonormal Basis and Spectral Transform
for: 本文提出了一种新的快速机器学习方法，利用了两种技术：Vector Embedding on Orthonormal Basis（VEOB）和Spectral Transform（ST）。
methods: 本方法使用了Singular Value Decomposition（SVD）技术计算向量基和投影坐标，从而提高了嵌入空间中的距离测量，并且可以压缩数据，保留最大的协方差 projection vectors。 ST 方法则将短 vectors 序列转换为спектраль空间，通过应用Discrete Cosine Transform（DCT）和选择最重要的组件，可以简化长 vectors 序列的处理。
results: 本文通过word embedding、text chunk embedding和image embedding的示例，在Julia语言中实现了一个向量数据库。它还 investigate了无监督学习和监督学习，以及处理大量数据的策略。Abstract
This paper presents a novel fast machine learning method that leverages two techniques: Vector Embedding on Orthonormal Basis (VEOB) and Spectral Transform (ST). The VEOB converts the original data encoding into a vector embedding with coordinates projected onto orthonormal bases. The Singular Value Decomposition (SVD) technique is used to calculate the vector basis and projection coordinates, leading to an enhanced distance measurement in the embedding space and facilitating data compression by preserving the projection vectors associated with the largest singular values. On the other hand, ST transforms sequence of vector data into spectral space. By applying the Discrete Cosine Transform (DCT) and selecting the most significant components, it streamlines the handling of lengthy vector sequences. The paper provides examples of word embedding, text chunk embedding, and image embedding, implemented in Julia language with a vector database. It also investigates unsupervised learning and supervised learning using this method, along with strategies for handling large data volumes.
摘要
这篇论文提出了一种新的快速机器学习方法，该方法利用了两种技术：向量嵌入在正交基（VEOB）和 спектраль转换（ST）。VEOB将原始数据编码转换为一个向量嵌入，其坐标被 проек到正交基上。使用SVD技术计算向量基和投影坐标，从而提高了嵌入空间中的距离测量，并且可以压缩数据，保留投影向量与最大特征值相关的 projet 矢量。一方面，ST将序列化的向量数据转换为 спектраль空间。通过应用DCT和选择最有价值的组件，可以简化长向量序列的处理。文章提供了word嵌入、文本块嵌入和图像嵌入的例子，实现在Julia语言中的一个向量数据库。它还 investigate了无监督学习和监督学习，以及处理大量数据的策略。
A general learning scheme for classical and quantum Ising machines
results: 实验结果表明，该学习模型在训练和执行上具有新的可能性，特别是在量子领域，量子资源被用于模型的执行和训练，提供了一个有前途的量子机器学习perspective。Abstract
An Ising machine is any hardware specifically designed for finding the ground state of the Ising model. Relevant examples are coherent Ising machines and quantum annealers. In this paper, we propose a new machine learning model that is based on the Ising structure and can be efficiently trained using gradient descent. We provide a mathematical characterization of the training process, which is based upon optimizing a loss function whose partial derivatives are not explicitly calculated but estimated by the Ising machine itself. Moreover, we present some experimental results on the training and execution of the proposed learning model. These results point out new possibilities offered by Ising machines for different learning tasks. In particular, in the quantum realm, the quantum resources are used for both the execution and the training of the model, providing a promising perspective in quantum machine learning.
摘要
一种Ising机器是专门设计用于找到Ising模型的稳定状态的硬件。相关的例子包括协调Isimg machine和量子气化器。在这篇论文中，我们提出一种基于Ising结构的新的机器学习模型，可以通过梯度下降方法高效地训练。我们提供了一个数学 caracterization of the training process，该过程基于优化一个损失函数的 partial derivatives不是直接计算出来，而是由Ising机器自己估算出来。此外，我们还提供了一些实验结果，证明了我们的学习模型在不同的任务上的应用前景。特别是在量子领域，量子资源被用于模型的执行和训练，提供了一个有前途的Perspective in quantum machine learning。
State-Action Similarity-Based Representations for Off-Policy Evaluation
results: 我们的实验结果显示，使用我们的状态动作表示方法可以提高 FQE 的数据效能，降低 OPE 错误值，并在不同的分布变换下保持 FQE 的稳定性。此外，我们还发现其他状态动作相似度度量无法表示评估策略的动作值函数，而我们的状态动作表示方法可以减少 FQE 中的数据误差。Abstract
In reinforcement learning, off-policy evaluation (OPE) is the problem of estimating the expected return of an evaluation policy given a fixed dataset that was collected by running one or more different policies. One of the more empirically successful algorithms for OPE has been the fitted q-evaluation (FQE) algorithm that uses temporal difference updates to learn an action-value function, which is then used to estimate the expected return of the evaluation policy. Typically, the original fixed dataset is fed directly into FQE to learn the action-value function of the evaluation policy. Instead, in this paper, we seek to enhance the data-efficiency of FQE by first transforming the fixed dataset using a learned encoder, and then feeding the transformed dataset into FQE. To learn such an encoder, we introduce an OPE-tailored state-action behavioral similarity metric, and use this metric and the fixed dataset to learn an encoder that models this metric. Theoretically, we show that this metric allows us to bound the error in the resulting OPE estimate. Empirically, we show that other state-action similarity metrics lead to representations that cannot represent the action-value function of the evaluation policy, and that our state-action representation method boosts the data-efficiency of FQE and lowers OPE error relative to other OPE-based representation learning methods on challenging OPE tasks. We also empirically show that the learned representations significantly mitigate divergence of FQE under varying distribution shifts. Our code is available here: https://github.com/Badger-RL/ROPE.
摘要
在强化学习中，评估策略（OPE）是计算评估策略所预期的返回的问题。一种更加实验成功的算法是适应q评估（FQE）算法，它使用时间差更新来学习一个动作价值函数，并使用这个函数来估计评估策略的返回。通常，原始固定数据集直接 fed 到 FQE 来学习评估策略的动作价值函数。而在这篇论文中，我们尝试使用一个学习的编码器来增强 FQE 的数据效率。我们引入一个适应 OPE 的状态动作行为相似度 metric，并使用这个 metric 和固定数据集来学习一个编码器，该编码器模型了这个 metric。理论上，我们证明这个 metric 可以约束 OPE 估计的错误。实验上，我们发现其他状态动作相似度 metric 导致的表示无法表示评估策略的动作价值函数，而我们的状态动作表示方法可以提高 FQE 的数据效率，相比其他基于 OPE 的表示学习方法。我们还发现学习的表示可以有效地减少 FQE 在不同分布下的急剧异常。我们的代码可以在以下链接中找到：https://github.com/Badger-RL/ROPE。
results: 研究发现，利用知道的分析物质含量来修正基线可以提高分析和量化结果的准确性，并且在两个近红外数据集上都有良好的性能。Abstract
Spectroscopic measurements can show distorted spectra shapes arising from a mixture of absorbing and scattering contributions. These distortions (or baselines) often manifest themselves as non-constant offsets or low-frequency oscillations. As a result, these baselines can adversely affect analytical and quantitative results. Baseline correction is an umbrella term where one applies pre-processing methods to obtain baseline spectra (the unwanted distortions) and then remove the distortions by differencing. However, current state-of-the art baseline correction methods do not utilize analyte concentrations even if they are available, or even if they contribute significantly to the observed spectral variability. We examine a class of state-of-the-art methods (penalized baseline correction) and modify them such that they can accommodate a priori analyte concentration such that prediction can be enhanced. Performance will be access on two near infra-red data sets across both classical penalized baseline correction methods (without analyte information) and modified penalized baseline correction methods (leveraging analyte information).
摘要
We examine a class of state-of-the-art baseline correction methods (penalized baseline correction) and modify them to accommodate a priori analyte concentration information. By leveraging this information, we can enhance prediction performance on two near infra-red data sets. In comparison to classical penalized baseline correction methods (without analyte information), our modified methods demonstrate improved performance.
Addressing GAN Training Instabilities via Tunable Classification Losses
paper_authors: Monica Welfert, Gowtham R. Kurri, Kyle Otstot, Lalitha Sankar
For: 该论文旨在提出一种基于生成对抗网络（GAN）的数据生成方法，使得生成的数据具有正式的保证。* Methods: 该论文使用类probability估计（CPE）损失函数来重新定义GAN的价值函数，并证明CPE损失GAN与$f$-GAN具有两种对应关系。此外，该论文还证明所有对称的$f$-散度都有相同的减少性。* Results: 在finite sample和模型容量下，该论文定义和获得估计和泛化错误的上下限。特别是，对于$\alpha$-GANs，该论文使用$\alpha$-损失函数，一个可调的CPE损失函数，并证明其在训练稳定性方面具有优越性。此外，该论文还引入了一种 dual-objective GAN，以解决GAN训练不稳定性问题。Abstract
Generative adversarial networks (GANs), modeled as a zero-sum game between a generator (G) and a discriminator (D), allow generating synthetic data with formal guarantees. Noting that D is a classifier, we begin by reformulating the GAN value function using class probability estimation (CPE) losses. We prove a two-way correspondence between CPE loss GANs and $f$-GANs which minimize $f$-divergences. We also show that all symmetric $f$-divergences are equivalent in convergence. In the finite sample and model capacity setting, we define and obtain bounds on estimation and generalization errors. We specialize these results to $\alpha$-GANs, defined using $\alpha$-loss, a tunable CPE loss family parametrized by $\alpha\in(0,\infty]$. We next introduce a class of dual-objective GANs to address training instabilities of GANs by modeling each player's objective using $\alpha$-loss to obtain $(\alpha_D,\alpha_G)$-GANs. We show that the resulting non-zero sum game simplifies to minimizing an $f$-divergence under appropriate conditions on $(\alpha_D,\alpha_G)$. Generalizing this dual-objective formulation using CPE losses, we define and obtain upper bounds on an appropriately defined estimation error. Finally, we highlight the value of tuning $(\alpha_D,\alpha_G)$ in alleviating training instabilities for the synthetic 2D Gaussian mixture ring as well as the large publicly available Celeb-A and LSUN Classroom image datasets.
摘要
生成敌战网络（GAN），模型为零SUM游戏中的生成器（G）和分类器（D）之间的对抗，可以生成具有正式保证的 sintetic 数据。注意到D是一个分类器，我们开始通过类型概率估计（CPE）损失函数来重新定义GAN的值函数。我们证明了CPE损失函数和$f$-GANs之间的双向对应关系，以及所有对称的$f$-散度都是相同的整合。在finite sample和模型容量设置下，我们定义和获得估计和泛化错误的上下界。我们特殊化这些结果到$\alpha$-GANs中，defined using $\alpha$-loss，一个可调CPE损失函数中的$\alpha\in(0,\infty]$.我们接着引入一个 dual-objective GAN来解决GAN的训练不稳定性，通过对每个玩家的目标使用$\alpha$-loss来获得$(\alpha_D,\alpha_G)$-GANs。我们证明了这个非零SUM游戏可以通过适当的条件来简化为$f$-散度的最小化。通过扩展这种双对象形式，我们定义和获得一个相应的估计错误的上限。最后，我们强调了在Synthetic 2D Gaussian mixture ring和大规模公共可用的Celeb-A和LSUN Classroom图像 dataset上调整($\alpha_D,\alpha_G)$的价值，以适应训练不稳定性。
results: 研究结果显示，使用开源的方法可以实现更好的 globel warming potential 和强度之间的交易，比现有业务实践更有利。Abstract
Eight percent of global carbon dioxide emissions can be attributed to the production of cement, the main component of concrete, which is also the dominant source of CO2 emissions in the construction of data centers. The discovery of lower-carbon concrete formulae is therefore of high significance for sustainability. However, experimenting with new concrete formulae is time consuming and labor intensive, as one usually has to wait to record the concrete's 28-day compressive strength, a quantity whose measurement can by its definition not be accelerated. This provides an opportunity for experimental design methodology like Bayesian Optimization (BO) to accelerate the search for strong and sustainable concrete formulae. Herein, we 1) propose modeling steps that make concrete strength amenable to be predicted accurately by a Gaussian process model with relatively few measurements, 2) formulate the search for sustainable concrete as a multi-objective optimization problem, and 3) leverage the proposed model to carry out multi-objective BO with real-world strength measurements of the algorithmically proposed mixes. Our experimental results show improved trade-offs between the mixtures' global warming potential (GWP) and their associated compressive strengths, compared to mixes based on current industry practices. Our methods are open-sourced at github.com/facebookresearch/SustainableConcrete.
摘要
全球碳排放的8%可以追溯到混凝土的生产，混凝土也是数据中心建设中主要的CO2排放来源。发现更低碳排放混凝土 формула的发现对可持续发展有着重要意义。然而，尝试新的混凝土 формула可以是时间占用和人力消耗的，因为一般需要等待28天压缩强度的测量，这个测量不能加速。这提供了 Bayesian 优化（BO）实验方法的机会，以加速搜索具有高强度和可持续的混凝土 формула。我们的方法包括：1. 模型步骤，使得混凝土强度可以准确预测，只需要 relativelly few 测量。2. 将寻找可持续的混凝土形式化为多目标优化问题。3. 利用我们提出的模型，通过实际测量 Algorithmically 提出的混凝土的强度，进行多目标 BO。我们的实验结果表明，我们的方法可以相比于现有行业实践，提高混凝土的全球温室效应（GWP）和压缩强度之间的交换。我们的方法在 GitHub 上公开发布，请参考 。
results: 我们的提议的 ESCFR 可以成功地解决很多治疗选择偏见问题，并在比较state-of-the-art方法时显示出显著的优异性。Abstract
Estimating conditional average treatment effect from observational data is highly challenging due to the existence of treatment selection bias. Prevalent methods mitigate this issue by aligning distributions of different treatment groups in the latent space. However, there are two critical problems that these methods fail to address: (1) mini-batch sampling effects (MSE), which causes misalignment in non-ideal mini-batches with outcome imbalance and outliers; (2) unobserved confounder effects (UCE), which results in inaccurate discrepancy calculation due to the neglect of unobserved confounders. To tackle these problems, we propose a principled approach named Entire Space CounterFactual Regression (ESCFR), which is a new take on optimal transport in the context of causality. Specifically, based on the framework of stochastic optimal transport, we propose a relaxed mass-preserving regularizer to address the MSE issue and design a proximal factual outcome regularizer to handle the UCE issue. Extensive experiments demonstrate that our proposed ESCFR can successfully tackle the treatment selection bias and achieve significantly better performance than state-of-the-art methods.
摘要
Estimating conditional average treatment effect from observational data is extremely difficult due to the existence of treatment selection bias. Prevalent methods mitigate this issue by aligning the distributions of different treatment groups in the latent space. However, there are two critical problems that these methods fail to address: (1) mini-batch sampling effects (MSE), which causes misalignment in non-ideal mini-batches with outcome imbalance and outliers; (2) unobserved confounder effects (UCE), which results in inaccurate discrepancy calculation due to the neglect of unobserved confounders. To tackle these problems, we propose a principled approach named Entire Space CounterFactual Regression (ESCFR), which is a new take on optimal transport in the context of causality. Specifically, based on the framework of stochastic optimal transport, we propose a relaxed mass-preserving regularizer to address the MSE issue and design a proximal factual outcome regularizer to handle the UCE issue. Extensive experiments demonstrate that our proposed ESCFR can successfully tackle the treatment selection bias and achieve significantly better performance than state-of-the-art methods.Here is the translation of the text into Traditional Chinese:Estimating conditional average treatment effect from observational data is extremely difficult due to the existence of treatment selection bias. Prevalent methods mitigate this issue by aligning the distributions of different treatment groups in the latent space. However, there are two critical problems that these methods fail to address: (1) mini-batch sampling effects (MSE), which causes misalignment in non-ideal mini-batches with outcome imbalance and outliers; (2) unobserved confounder effects (UCE), which results in inaccurate discrepancy calculation due to the neglect of unobserved confounders. To tackle these problems, we propose a principled approach named Entire Space CounterFactual Regression (ESCFR), which is a new take on optimal transport in the context of causality. Specifically, based on the framework of stochastic optimal transport, we propose a relaxed mass-preserving regularizer to address the MSE issue and design a proximal factual outcome regularizer to handle the UCE issue. Extensive experiments demonstrate that our proposed ESCFR can successfully tackle the treatment selection bias and achieve significantly better performance than state-of-the-art methods.
Entity Embeddings : Perspectives Towards an Omni-Modality Era for Large Language Models
results: 该论文预测了这种结构的潜在应用和挑战。Abstract
Large Language Models (LLMs) are evolving to integrate multiple modalities, such as text, image, and audio into a unified linguistic space. We envision a future direction based on this framework where conceptual entities defined in sequences of text can also be imagined as modalities. Such a formulation has the potential to overcome the cognitive and computational limitations of current models. Several illustrative examples of such potential implicit modalities are given. Along with vast promises of the hypothesized structure, expected challenges are discussed as well.
摘要
大型语言模型（LLM）在演化中将多Modalities，如文本、图像和音频 integrate into a unified linguistic space。我们可以想像将概念实体定义为文本序列中的一部分，也可以被视为模式。这种概念的形式化有很多潜在的应用前景，并且可以超越当前模型的认知和计算限制。我们给出了一些示例，以及预期的挑战。
Structured Semidefinite Programming for Recovering Structured Preconditioners
paper_authors: Arun Jambulapati, Jerry Li, Christopher Musco, Kirankumar Shiragur, Aaron Sidford, Kevin Tian
for: 这种框架用于解决线性系统的约似优化预处理问题。
methods: 使用新的核心矩阵方法和矩阵解决问题来解决这种问题。
results: 得到了改进的运行时间，比如$\widetilde{O}(\text{nnz}(\mathbf{K}) \cdot \text{poly}(\kappa^\star,\epsilon^{-1}))$和$\widetilde{O}(d^2)$。Abstract
We develop a general framework for finding approximately-optimal preconditioners for solving linear systems. Leveraging this framework we obtain improved runtimes for fundamental preconditioning and linear system solving problems including the following. We give an algorithm which, given positive definite $\mathbf{K} \in \mathbb{R}^{d \times d}$ with $\mathrm{nnz}(\mathbf{K})$ nonzero entries, computes an $\epsilon$-optimal diagonal preconditioner in time $\widetilde{O}(\mathrm{nnz}(\mathbf{K}) \cdot \mathrm{poly}(\kappa^\star,\epsilon^{-1}))$, where $\kappa^\star$ is the optimal condition number of the rescaled matrix. We give an algorithm which, given $\mathbf{M} \in \mathbb{R}^{d \times d}$ that is either the pseudoinverse of a graph Laplacian matrix or a constant spectral approximation of one, solves linear systems in $\mathbf{M}$ in $\widetilde{O}(d^2)$ time. Our diagonal preconditioning results improve state-of-the-art runtimes of $\Omega(d^{3.5})$ attained by general-purpose semidefinite programming, and our solvers improve state-of-the-art runtimes of $\Omega(d^{\omega})$ where $\omega > 2.3$ is the current matrix multiplication constant. We attain our results via new algorithms for a class of semidefinite programs (SDPs) we call matrix-dictionary approximation SDPs, which we leverage to solve an associated problem we call matrix-dictionary recovery.
摘要
我们开发了一个通用框架，用于找到近似优化的预Conditioner，以解决线性系统。利用这个框架，我们得到了改进的运行时间 для基本的预Conditioning和线性系统解决问题，包括以下几个。我们提供了一个算法， Given positive definite $\mathbf{K} \in \mathbb{R}^{d \times d}$ with $\mathrm{nnz}(\mathbf{K})$ nonzero entries, computes an $\epsilon$-optimal diagonal预Conditioner in time $\widetilde{O}(\mathrm{nnz}(\mathbf{K}) \cdot \mathrm{poly}(\kappa^\star,\epsilon^{-1}))$, where $\kappa^\star$ is the optimal condition number of the rescaled matrix.我们提供了一个算法， Given $\mathbf{M} \in \mathbb{R}^{d \times d}$ that is either the pseudoinverse of a graph Laplacian matrix or a constant spectral approximation of one, solves linear systems in $\mathbf{M}$ in $\widetilde{O}(d^2)$ time.我们的diagonal预Conditioning结果提高了现有的semidefinite programming的运行时间，从而实现了$\Omega(d^{3.5})$的性能。我们的解决方案提高了现有的运行时间，达到了$\Omega(d^{\omega})$，其中 $\omega > 2.3$ 是当前的矩阵乘法常数。我们通过新的semidefinite programs（SDPs）的算法，称为matrix-dictionary approximation SDPs，来解决一个 associate problem，称为matrix-dictionary recovery。
Guided Data Augmentation for Offline Reinforcement Learning and Imitation Learning
results: 我们对physical robot Soccer任务和simulated D4RL导航任务、simulated autonomous driving任务和simulated Soccer任务进行了实验，发现GuDA可以学习从一小量可能不优质的示例中，并大幅超越随机生成的DA策略。Abstract
Learning from demonstration (LfD) is a popular technique that uses expert demonstrations to learn robot control policies. However, the difficulty in acquiring expert-quality demonstrations limits the applicability of LfD methods: real-world data collection is often costly, and the quality of the demonstrations depends greatly on the demonstrator's abilities and safety concerns. A number of works have leveraged data augmentation (DA) to inexpensively generate additional demonstration data, but most DA works generate augmented data in a random fashion and ultimately produce highly suboptimal data. In this work, we propose Guided Data Augmentation (GuDA), a human-guided DA framework that generates expert-quality augmented data. The key insight of GuDA is that while it may be difficult to demonstrate the sequence of actions required to produce expert data, a user can often easily identify when an augmented trajectory segment represents task progress. Thus, the user can impose a series of simple rules on the DA process to automatically generate augmented samples that approximate expert behavior. To extract a policy from GuDA, we use off-the-shelf offline reinforcement learning and behavior cloning algorithms. We evaluate GuDA on a physical robot soccer task as well as simulated D4RL navigation tasks, a simulated autonomous driving task, and a simulated soccer task. Empirically, we find that GuDA enables learning from a small set of potentially suboptimal demonstrations and substantially outperforms a DA strategy that samples augmented data randomly.
摘要
学习示例（LfD）是一种广泛使用的技术，通过专家示例学习机器人控制策略。然而，获得专家质量示例的困难限制了LfD方法的应用范围：现实世界数据收集常常成本高昂，示例制定者的能力和安全问题具有很大的影响。许多工作使用数据扩展（DA）来生成更多的示例数据，但大多数DA工作生成扩展数据的方式是随机的，最终生成低质量的数据。在这种情况下，我们提出了指导数据扩展（GuDA），一种人类指导的DA框架，可以生成专家质量的扩展数据。GuDA的关键想法是，虽然可能困难示出完整的任务执行序列，但用户可以轻松地判断扩展 trajectory 段是否表示任务进步。因此，用户可以对 DA 过程进行一些简单的规则，自动生成扩展样本，approxime 专家行为。为提取策略，我们使用现有的离线 reinforcement learning 和行为复制算法。我们对 physical robot 足球任务、simulated D4RL 导航任务、simulated autonomous driving 任务和 simulated 足球任务 进行了评估。 empirically，我们发现 GuDA 可以从小型可能不优质的示例中学习，并substantially 超过随机扩展数据的DA策略。
$α$-Mutual Information: A Tunable Privacy Measure for Privacy Protection in Data Sharing
methods: 我们使用了一种通用的扭曲基本机制， manipulate the original data to offer privacy protection。该扭曲度量根据具体的实验数据结构进行确定。
results: 我们通过实验证明了α-私钥信息的适用性，并证明了我们的方法可以在不同的性能维度上妥协隐私和实用性。此外，我们还分析了攻击者获取私人数据的边 información的后果，并证明了我们的适应性比现有技术更高。Abstract
This paper adopts Arimoto's $\alpha$-Mutual Information as a tunable privacy measure, in a privacy-preserving data release setting that aims to prevent disclosing private data to adversaries. By fine-tuning the privacy metric, we demonstrate that our approach yields superior models that effectively thwart attackers across various performance dimensions. We formulate a general distortion-based mechanism that manipulates the original data to offer privacy protection. The distortion metrics are determined according to the data structure of a specific experiment. We confront the problem expressed in the formulation by employing a general adversarial deep learning framework that consists of a releaser and an adversary, trained with opposite goals. This study conducts empirical experiments on images and time-series data to verify the functionality of $\alpha$-Mutual Information. We evaluate the privacy-utility trade-off of customized models and compare them to mutual information as the baseline measure. Finally, we analyze the consequence of an attacker's access to side information about private data and witness that adapting the privacy measure results in a more refined model than the state-of-the-art in terms of resiliency against side information.
摘要
Note: The text has been translated into Simplified Chinese, which is the standard writing system used in mainland China. The translation may not be perfect, and some nuances or idioms may be lost in translation.
paper_authors: Sáez-Maldonado Francisco Javier, Maroñas Juan, Hernández-Lobato Daniel
for: 本研究旨在探讨增强 Gaussian Processes (GP) 的扩展，提高 GP 的灵活性。
methods: 本文提出了一种新的扩展方法，即 Deep Transformed Gaussian Processes (DTGP)，它是通过堆叠多个噪声过程层来增强 GP 的灵活性。
results: 实验表明，DTGP 可以在多个回归数据集中实现好的扩展性和性能。Abstract
Transformed Gaussian Processes (TGPs) are stochastic processes specified by transforming samples from the joint distribution from a prior process (typically a GP) using an invertible transformation; increasing the flexibility of the base process. Furthermore, they achieve competitive results compared with Deep Gaussian Processes (DGPs), which are another generalization constructed by a hierarchical concatenation of GPs. In this work, we propose a generalization of TGPs named Deep Transformed Gaussian Processes (DTGPs), which follows the trend of concatenating layers of stochastic processes. More precisely, we obtain a multi-layer model in which each layer is a TGP. This generalization implies an increment of flexibility with respect to both TGPs and DGPs. Exact inference in such a model is intractable. However, we show that one can use variational inference to approximate the required computations yielding a straightforward extension of the popular DSVI inference algorithm Salimbeni et al (2017). The experiments conducted evaluate the proposed novel DTGPs in multiple regression datasets, achieving good scalability and performance.
摘要
transformed Gaussian 进程（TGP）是一种 Stochastic 过程，其 Specified by transforming samples from the joint distribution of a prior process (usually a GP) using an invertible transformation; increasing the flexibility of the base process. Furthermore, they achieve competitive results compared with deep Gaussian processes (DGPs), which are another generalization constructed by a hierarchical concatenation of GPs. In this work, we propose a generalization of TGPs named deep transformed Gaussian processes (DTGPs), which follows the trend of concatenating layers of stochastic processes. More precisely, we obtain a multi-layer model in which each layer is a TGP. This generalization implies an increase in flexibility with respect to both TGPs and DGPs. Exact inference in such a model is intractable. However, we show that one can use variational inference to approximate the required computations, yielding a straightforward extension of the popular DSVI inference algorithm (Salimbeni et al., 2017). The experiments conducted evaluate the proposed novel DTGPs in multiple regression datasets, achieving good scalability and performance.Note: Some of the technical terms in the original text, such as "Gaussian processes" and "variational inference," may not have direct translations in Simplified Chinese. In such cases, I have used the most common translations available in the literature, but the reader may need to consult a more specialized dictionary or reference for a more precise translation.
One Model Fits All: Cross-Region Taxi-Demand Forecasting
results: 实验结果表明，提案的系统能够准确预测出行需求，包括在未经见过的区域。这显示了该系统在优化出行服务和提高交通效率的潜力。Abstract
The growing demand for ride-hailing services has led to an increasing need for accurate taxi demand prediction. Existing systems are limited to specific regions, lacking generalizability to unseen areas. This paper presents a novel taxi demand forecasting system that leverages a graph neural network to capture spatial dependencies and patterns in urban environments. Additionally, the proposed system employs a region-neutral approach, enabling it to train a model that can be applied to any region, including unseen regions. To achieve this, the framework incorporates the power of Variational Autoencoder to disentangle the input features into region-specific and region-neutral components. The region-neutral features facilitate cross-region taxi demand predictions, allowing the model to generalize well across different urban areas. Experimental results demonstrate the effectiveness of the proposed system in accurately forecasting taxi demand, even in previously unobserved regions, thus showcasing its potential for optimizing taxi services and improving transportation efficiency on a broader scale.
摘要
随着乘车需求的增长，需求预测已成为了ride-hailing服务的紧迫需求。现有的系统受限于特定地区，缺乏对未见地区的泛化能力。这篇论文提出了一种新的出租车需求预测系统，利用图 neural network 捕捉城市环境中的空间依赖关系和模式。此外，提出的系统采用了地域中性的方法，使得其可以训练可应用于任何地区，包括未见地区的模型。为达到这一目标，框架具有Variational Autoencoder 的力量，压缩输入特征成地域特定和地域中性组成部分。地域中性特征使得出租车需求预测可以在不同的城市区域进行跨地区预测，使模型能够在不同的城市区域中具有泛化能力。实验结果表明，提出的系统可以准确预测出租车需求，甚至在未见地区进行预测，从而展示其在优化出租车服务和改善交通效率的潜力。
Robustness of Algorithms for Causal Structure Learning to Hyperparameter Choice
paper_authors: Damian Machlanski, Spyridon Samothrakis, Paul Clarke
for: 本研究旨在探讨隐藏参数在结构学习中的影响，以及如何选择最佳的隐藏参数来提高结构学习性能。
methods: 本研究采用了一些经典的结构学习算法，并对这些算法进行了融合调参。
results: 研究发现，隐藏参数的选择在集成设置中具有很大的影响，可以导致分析者选择不适合自己数据的算法，从而影响结构学习性能。Abstract
Hyperparameters play a critical role in machine learning. Hyperparameter tuning can make the difference between state-of-the-art and poor prediction performance for any algorithm, but it is particularly challenging for structure learning due to its unsupervised nature. As a result, hyperparameter tuning is often neglected in favour of using the default values provided by a particular implementation of an algorithm. While there have been numerous studies on performance evaluation of causal discovery algorithms, how hyperparameters affect individual algorithms, as well as the choice of the best algorithm for a specific problem, has not been studied in depth before. This work addresses this gap by investigating the influence of hyperparameters on causal structure learning tasks. Specifically, we perform an empirical evaluation of hyperparameter selection for some seminal learning algorithms on datasets of varying levels of complexity. We find that, while the choice of algorithm remains crucial to obtaining state-of-the-art performance, hyperparameter selection in ensemble settings strongly influences the choice of algorithm, in that a poor choice of hyperparameters can lead to analysts using algorithms which do not give state-of-the-art performance for their data.
摘要
Model-free Posterior Sampling via Learning Rate Randomization
paper_authors: Daniil Tiapkin, Denis Belomestny, Daniele Calandriello, Eric Moulines, Remi Munos, Alexey Naumov, Pierre Perrault, Michal Valko, Pierre Menard
results: RandQL 在 tabular MDPs 和 metric state-action space 中均可以 achieve regret bound of order $\widetilde{\mathcal{O}(\sqrt{H^{5}SAT})$，并且在 empirical study 中与 existing approaches 比较，表现更好。Abstract
In this paper, we introduce Randomized Q-learning (RandQL), a novel randomized model-free algorithm for regret minimization in episodic Markov Decision Processes (MDPs). To the best of our knowledge, RandQL is the first tractable model-free posterior sampling-based algorithm. We analyze the performance of RandQL in both tabular and non-tabular metric space settings. In tabular MDPs, RandQL achieves a regret bound of order $\widetilde{\mathcal{O}(\sqrt{H^{5}SAT})$, where $H$ is the planning horizon, $S$ is the number of states, $A$ is the number of actions, and $T$ is the number of episodes. For a metric state-action space, RandQL enjoys a regret bound of order $\widetilde{\mathcal{O}(H^{5/2} T^{(d_z+1)/(d_z+2)})$, where $d_z$ denotes the zooming dimension. Notably, RandQL achieves optimistic exploration without using bonuses, relying instead on a novel idea of learning rate randomization. Our empirical study shows that RandQL outperforms existing approaches on baseline exploration environments.
摘要
在这篇论文中，我们介绍了Randomized Q-learning（RandQL），一种新的随机化无模型算法，用于减少偏差在集束Markov决策过程（MDP）中的 regret。我们知道，RandQL是首个可追踪的无模型 posterior sampling-based algorithm。我们分析了RandQL在标准MDP和非标准 metric space中的性能。在标准MDP中，RandQL的 regret bound为 $\widetilde{\mathcal{O}(\sqrt{H^{5}SAT})$，其中 $H$ 是规划 horizion， $S$ 是状态数量， $A$ 是动作数量， $T$ 是集束数量。在 metric state-action space 中，RandQL 的 regret bound为 $\widetilde{\mathcal{O}(H^{5/2} T^{(d_z+1)/(d_z+2)})$，其中 $d_z$ 表示 zooming 维度。值得注意的是，RandQL 实现了无奖诱导的探索，不使用奖励，而是通过一种新的学习率随机化的想法。我们的实验研究表明，RandQL 在基eline探索环境中表现出色。
Enhancing Enterprise Network Security: Comparing Machine-Level and Process-Level Analysis for Dynamic Malware Detection
results: 对比前一代STATE-OF-THE-ART方法，本研究的提议模型具有较高的检测精度，具体来说，检测精度提高约20.12%，而false positive率在0.1左右。Abstract
Analysing malware is important to understand how malicious software works and to develop appropriate detection and prevention methods. Dynamic analysis can overcome evasion techniques commonly used to bypass static analysis and provide insights into malware runtime activities. Much research on dynamic analysis focused on investigating machine-level information (e.g., CPU, memory, network usage) to identify whether a machine is running malicious activities. A malicious machine does not necessarily mean all running processes on the machine are also malicious. If we can isolate the malicious process instead of isolating the whole machine, we could kill the malicious process, and the machine can keep doing its job. Another challenge dynamic malware detection research faces is that the samples are executed in one machine without any background applications running. It is unrealistic as a computer typically runs many benign (background) applications when a malware incident happens. Our experiment with machine-level data shows that the existence of background applications decreases previous state-of-the-art accuracy by about 20.12% on average. We also proposed a process-level Recurrent Neural Network (RNN)-based detection model. Our proposed model performs better than the machine-level detection model; 0.049 increase in detection rate and a false-positive rate below 0.1.
摘要
Proportional Fairness in Clustering: A Social Choice Perspective
results: 这篇论文显示任何归一化都能同时满足分配公平性定义和个人公平性定义，以及更强的多赢者代表性定义。此外，弱 пропорциональ性定义也能导致更强的多赢者代表性定义的近似值。Abstract
We study the proportional clustering problem of Chen et al. [ICML'19] and relate it to the area of multiwinner voting in computational social choice. We show that any clustering satisfying a weak proportionality notion of Brill and Peters [EC'23] simultaneously obtains the best known approximations to the proportional fairness notion of Chen et al. [ICML'19], but also to individual fairness [Jung et al., FORC'20] and the "core" [Li et al. ICML'21]. In fact, we show that any approximation to proportional fairness is also an approximation to individual fairness and vice versa. Finally, we also study stronger notions of proportional representation, in which deviations do not only happen to single, but multiple candidate centers, and show that stronger proportionality notions of Brill and Peters [EC'23] imply approximations to these stronger guarantees.
摘要
我们研究陈等人的著作中的协Relative Clustering问题([ICML'19])，并与计算社会选择领域的多赢者投票问题相关。我们显示任何满足Brill和Peter斯([EC'23])的弱 пропорциональ性定义，同时也能获得Chen等人的最佳知道的变分概念([ICML'19])、个体公平([Jung等人，FORC'20])以及"核心"([Li等人，ICML'21])的最佳近似。事实上，任何对 proportional fairness 的近似也是对个体公平的近似，并且vice versa。 finally，我们还研究了更强的多个候选者代表性定义，在多个候选者中偏移不仅发生在单个中心，而是在多个中心上，并证明Brill和Peter斯([EC'23])的更强的 proportionality 定义能够导致这些更强的保证。
Sample Complexity Bounds for Score-Matching: Causal Discovery and Generative Modeling
results: 论文提出了对Score-matching基于 causal discovery方法的Recovering causal relationships的误差率的下界，假设得分函数的估计充分准确。此外，论文还分析了Score-matching估计在Score-based生成模型中的Upper bound。Abstract
This paper provides statistical sample complexity bounds for score-matching and its applications in causal discovery. We demonstrate that accurate estimation of the score function is achievable by training a standard deep ReLU neural network using stochastic gradient descent. We establish bounds on the error rate of recovering causal relationships using the score-matching-based causal discovery method of Rolland et al. [2022], assuming a sufficiently good estimation of the score function. Finally, we analyze the upper bound of score-matching estimation within the score-based generative modeling, which has been applied for causal discovery but is also of independent interest within the domain of generative models.
摘要
Note:* "score-matching" is translated as "分数匹配" (fēnzhèng píngchǎ)* "causal discovery" is translated as " causal discovery" ( causal discovery)* "score function" is translated as "分数函数" (fēnzhèng fúnción)* "standard deep ReLU neural network" is translated as "标准深度ReLU神经网络" (zhèngdé ReLU xīnnéirwàng)* "stochastic gradient descent" is translated as "随机梯度下降" (suìjī tiēdào xiàojiù)* "error rate" is translated as "错误率" (error rate)
A Global Multi-Unit Calibration as a Method for Large Scale IoT Particulate Matter Monitoring Systems Deployments
results: 测试campaign表明，当应用于不同的传感器时，这种方法的性能与现有的方法匹配，而且可以实现大量的准确气囊监测设备的投入。如果确认，这些结果表明，当得到了正确的准确测试法，可以在大量的网络设备上实现准确的气囊监测，并且可以减少长距离数据传输需求。此外，这种准确测试模型可以轻松地被嵌入到设备上，或者在边缘实现，以便实现个人曝露监测应用。Abstract
Scalable and effective calibration is a fundamental requirement for Low Cost Air Quality Monitoring Systems and will enable accurate and pervasive monitoring in cities. Suffering from environmental interferences and fabrication variance, these devices need to encompass sensors specific and complex calibration processes for reaching a sufficient accuracy to be deployed as indicative measurement devices in Air Quality (AQ) monitoring networks. Concept and sensor drift often force calibration process to be frequently repeated. These issues lead to unbearable calibration costs which denies their massive deployment when accuracy is a concern. In this work, We propose a zero transfer samples, global calibration methodology as a technological enabler for IoT AQ multisensory devices which relies on low cost Particulate Matter (PM) sensors. This methodology is based on field recorded responses from a limited number of IoT AQ multisensors units and machine learning concepts and can be universally applied to all units of the same type. A multi season test campaign shown that, when applied to different sensors, this methodology performances match those of state of the art methodology which requires to derive different calibration parameters for each different unit. If confirmed, these results show that, when properly derived, a global calibration law can be exploited for a large number of networked devices with dramatic cost reduction eventually allowing massive deployment of accurate IoT AQ monitoring devices. Furthermore, this calibration model could be easily embedded on board of the device or implemented on the edge allowing immediate access to accurate readings for personal exposure monitor applications as well as reducing long range data transfer needs.
摘要
低成本空气质量监测系统需要扩展可扩展的准确化，以实现精准和广泛的监测城市。由于环境干扰和制造变化，这些设备需要特定的感应器和复杂的准确化过程以达到足够的准确性，以便作为空气质量（AQ）监测网络的指示测量设备。概念和感应器偏移 часто导致准确化过程需要频繁重复。这些问题导致不可持续的准确化成本，这使得大规模部署变得不可能。在这种情况下，我们提议一种零传输样本、全球准确化方法，这种方法基于低成本 particulate matter（PM）感应器。这种方法基于场记录的响应，并且可以通过机器学习概念应用于所有类型的单元。一个多季度测试 campagne表明，当应用于不同的感应器时，这种方法的性能与当前的方法匹配，该方法需要为每个不同单元 derivation 不同的准确化参数。如果确认，这些结果表明，当正确地 derivation 全球准确化法则，可以在大量部署精准的 IoT AQ 监测设备。此外，这种准确化模型可以轻松地嵌入到设备上或实现在边缘，以便提供快速的准确阅读，用于个人曝露监测应用，以及减少长距离数据传输需求。
Transductive conformal inference with adaptive scores
For: This paper provides distribution-free guarantees for many machine learning tasks, specifically in the transductive setting where decisions are made on a test sample of new points.* Methods: The paper uses conformal inference, which is a fundamental and versatile tool that provides distribution-free guarantees. The paper also uses a P'olya urn model to describe the joint distribution of the conformal $p$-values, and establishes a concentration inequality for their empirical distribution function.* Results: The paper provides uniform, in-probability guarantees for two machine learning tasks of current interest: interval prediction for transductive transfer learning and novelty detection based on two-class classification. The results hold for arbitrary exchangeable scores, including adaptive ones that can use the covariates of the test+calibration samples at training stage for increased accuracy.Abstract
Conformal inference is a fundamental and versatile tool that provides distribution-free guarantees for many machine learning tasks. We consider the transductive setting, where decisions are made on a test sample of $m$ new points, giving rise to $m$ conformal $p$-values. {While classical results only concern their marginal distribution, we show that their joint distribution follows a P\'olya urn model, and establish a concentration inequality for their empirical distribution function.} The results hold for arbitrary exchangeable scores, including {\it adaptive} ones that can use the covariates of the test+calibration samples at training stage for increased accuracy. We demonstrate the usefulness of these theoretical results through uniform, in-probability guarantees for two machine learning tasks of current interest: interval prediction for transductive transfer learning and novelty detection based on two-class classification.
摘要
它们是一种基本和多方面的工具，提供不受分布限制的保证，用于许多机器学习任务。我们考虑了推论 setting，在一个测试样本中有 $m$ 个新点，从而生成 $m$ 个充分满足的 $p$-值。{而 classical 结果只关注它们的边缘分布，我们显示它们的联合分布遵循波尔雅urn模型，并证明它们的empirical distribution function具有减法不等式.}结果适用于任意兼容的分数，包括可适应的分数，可以在训练阶段使用测试样本和标准化样本的 covariates 进行更高的准确性。我们通过对两个现有的机器学习任务进行保证， namely interval prediction for transductive transfer learning和 noveldetection based on two-class classification，来证明这些理论结果的实用性。
Adversarial Anomaly Detection using Gaussian Priors and Nonlinear Anomaly Scores
results: 与现有工作相比，该论文在MITBIH Arrhythmia Database 上的 $F_1$ 分数从0.85提高到0.92，表明β-VAEGAN 可以更好地检测异常。Abstract
Anomaly detection in imbalanced datasets is a frequent and crucial problem, especially in the medical domain where retrieving and labeling irregularities is often expensive. By combining the generative stability of a $\beta$-variational autoencoder (VAE) with the discriminative strengths of generative adversarial networks (GANs), we propose a novel model, $\beta$-VAEGAN. We investigate methods for composing anomaly scores based on the discriminative and reconstructive capabilities of our model. Existing work focuses on linear combinations of these components to determine if data is anomalous. We advance existing work by training a kernelized support vector machine (SVM) on the respective error components to also consider nonlinear relationships. This improves anomaly detection performance, while allowing faster optimization. Lastly, we use the deviations from the Gaussian prior of $\beta$-VAEGAN to form a novel anomaly score component. In comparison to state-of-the-art work, we improve the $F_1$ score during anomaly detection from 0.85 to 0.92 on the widely used MITBIH Arrhythmia Database.
摘要
非常常见的异常检测问题在不均衡数据集中，特别是在医疗领域，因为检测和标注异常性往往是昂贵的。我们提出了一种新的模型，$\beta$-VAEGAN，通过结合$\beta$-variational autoencoder（VAE）的生成稳定性和生成敌对网络（GANs）的攻击力，以提高异常检测性能。我们研究了基于这两个组件的异常分数的组合方法，包括线性组合以及训练kernelized支持向量机（SVM）来考虑非线性关系。这些改进了异常检测性能，同时允许更快的优化。此外，我们还使用$\beta$-VAEGAN的偏差从拜尔分布来形成一种新的异常分数组件。与现有工作相比，我们在MITBIHArrhythmia数据库上提高了异常检测$F_1$分数从0.85提高到0.92。
Unveiling the Potential of Probabilistic Embeddings in Self-Supervised Learning
results: 研究发现, 在信息理论基础下, 增加一个压缩瓶颈可以明显提高外部样本探测能力, 但是同时可能导致表示的压缩和信息损失。Abstract
In recent years, self-supervised learning has played a pivotal role in advancing machine learning by allowing models to acquire meaningful representations from unlabeled data. An intriguing research avenue involves developing self-supervised models within an information-theoretic framework, but many studies often deviate from the stochasticity assumptions made when deriving their objectives. To gain deeper insights into this issue, we propose to explicitly model the representation with stochastic embeddings and assess their effects on performance, information compression and potential for out-of-distribution detection. From an information-theoretic perspective, we seek to investigate the impact of probabilistic modeling on the information bottleneck, shedding light on a trade-off between compression and preservation of information in both representation and loss space. Emphasizing the importance of distinguishing between these two spaces, we demonstrate how constraining one can affect the other, potentially leading to performance degradation. Moreover, our findings suggest that introducing an additional bottleneck in the loss space can significantly enhance the ability to detect out-of-distribution examples, only leveraging either representation features or the variance of their underlying distribution.
摘要
近年来，自适应学习已经在机器学习领域发挥了关键作用，允许模型从无标签数据中获得有意义的表示。一个吸引人的研究方向是在信息理论框架下开发自适应模型，但许多研究通常会背离在 derivation 目标时所做的随机性假设。为了更深入地了解这个问题，我们提议显式地模型表示中的随机嵌入，评估其对性能、信息压缩和可能出现在其他分布中的检测的影响。从信息理论的视角来看，我们希望 investigate 表示中的信息瓶颈，探讨其与损失空间之间的交互关系，以及在这两个空间之间是否存在负反馈的问题。我们发现，在损失空间中引入一个额外的瓶颈可以明显提高对于异常分布的检测，只需要使用表示特征或者对其分布的变异。
Lipschitz and Hölder Continuity in Reproducing Kernel Hilbert Spaces
for: investigate Lipschitz and H"older continuity in Reproducing Kernel Hilbert Spaces (RKHSs)
methods: provide sufficient conditions and collect related known results from the literature
results: new results on reproducing kernels inducing prescribed Lipschitz or H"older continuityAbstract
Reproducing kernel Hilbert spaces (RKHSs) are very important function spaces, playing an important role in machine learning, statistics, numerical analysis and pure mathematics. Since Lipschitz and H\"older continuity are important regularity properties, with many applications in interpolation, approximation and optimization problems, in this work we investigate these continuity notion in RKHSs. We provide several sufficient conditions as well as an in depth investigation of reproducing kernels inducing prescribed Lipschitz or H\"older continuity. Apart from new results, we also collect related known results from the literature, making the present work also a convenient reference on this topic.
摘要
<>将文本翻译成简化中文。<>复制kernel空间（RKHS）是非常重要的函数空间，在机器学习、统计、数值分析和纯 математи学中扮演着重要的角色。由于 lipschitz 和 holder 连续性是重要的规范性质，在 interpolate、approximation 和优化问题中具有广泛的应用，因此在这种工作中我们调查这些连续性观念在 RKHS 中。我们提供了多个足够条件以及对 reproduce kernel 引起的 lipschitz 或 holder 连续性进行深入调查。除了新的结果之外，我们还收集了相关的已知结果，使得现在的工作也成为了这个话题的便捷参考。
On kernel-based statistical learning in the mean field limit
results: 研究结果表明，在mean field limit中，empirical和无限样本解的 converges 以及相关的风险的 converges。这些结果为大规模问题提供了新的理论工具和洞察，同时也为统计学学习理论中的limit问题提供了新的形式。Abstract
In many applications of machine learning, a large number of variables are considered. Motivated by machine learning of interacting particle systems, we consider the situation when the number of input variables goes to infinity. First, we continue the recent investigation of the mean field limit of kernels and their reproducing kernel Hilbert spaces, completing the existing theory. Next, we provide results relevant for approximation with such kernels in the mean field limit, including a representer theorem. Finally, we use these kernels in the context of statistical learning in the mean field limit, focusing on Support Vector Machines. In particular, we show mean field convergence of empirical and infinite-sample solutions as well as the convergence of the corresponding risks. On the one hand, our results establish rigorous mean field limits in the context of kernel methods, providing new theoretical tools and insights for large-scale problems. On the other hand, our setting corresponds to a new form of limit of learning problems, which seems to have not been investigated yet in the statistical learning theory literature.
摘要
Many machine learning applications involve a large number of variables. 基于机器学习中的互动体系，我们考虑到输入变量的数量趋于无穷大。首先，我们继续推动渐近场限的kernel和它们的重现函数空间的研究，完善现有的理论。接着，我们提供用这些kernel进行近似的结果，包括一个表示定理。最后，我们使用这些kernel在mean field限下进行统计学学习，具体来说，我们展示了empirical和无限样本解的mean field收敛和相应的风险的收敛。我们的结果建立了机器学习中的mean field限，提供了新的理论工具和意见，用于处理大规模问题。另一方面，我们的设定对 statistical learning theory文献中没有被 investigate的一种新的限制问题形式，这种形式是mean field limit。
results: 该论文提出的方法可以提高梯度约束的精度，从而提高数据隐私的保护水平，同时也可以降低噪声的水平。Abstract
Recently, due to the popularity of deep neural networks and other methods whose training typically relies on the optimization of an objective function, and due to concerns for data privacy, there is a lot of interest in differentially private gradient descent methods. To achieve differential privacy guarantees with a minimum amount of noise, it is important to be able to bound precisely the sensitivity of the information which the participants will observe. In this study, we present a novel approach that mitigates the bias arising from traditional gradient clipping. By leveraging public information concerning the current global model and its location within the search domain, we can achieve improved gradient bounds, leading to enhanced sensitivity determinations and refined noise level adjustments. We extend the state of the art algorithms, present improved differential privacy guarantees requiring less noise and present an empirical evaluation.
摘要
最近，由于深度神经网络和其他方法的训练通常基于目标函数优化，以及数据隐私的关注，有很多关注在不同敏感度下进行梯度下降方法。为了保证数据隐私保障，需要准确地评估梯度下降中信息敏感度。在本研究中，我们提出了一种新的方法，用于减轻传统梯度裁剪所导致的偏见。我们利用公共信息，包括当前全球模型和其位置在搜索区域中，来实现更好的梯度 bound，从而提高敏感度评估和降低噪声水平。我们扩展了现有算法，提供更好的不同敏感度保障，并进行了实验评估。
Closing the Gap Between the Upper Bound and the Lower Bound of Adam’s Iteration Complexity
paper_authors: Bohan Wang, Jingwen Fu, Huishuai Zhang, Nanning Zheng, Wei Chen for: 这个论文是为了提供一种新的 Adam 优化算法的 convergenc guarantee，以便在不同的 hyperparameters 下实现更高的效率。methods: 这个论文使用了一种新的技术来处理积分和自适应学习率的杂糅，并将 Descent Lemma 中的首项转换为 gradients 的 norm，以获得更高的效率。results: 这个论文提出了一种新的 Adam 优化算法，其 convergenc guarantee 是基于 $L$-smooth condition 和 bounded noise variance assumption，并且适用于广泛的 hyperparameters。特别是，对于合适的 hyperparameters，这个算法可以实现更高的效率，并且可以 closing the gap между existing literature 中的 convergence guarantee 和实际性能。Abstract
Recently, Arjevani et al. [1] established a lower bound of iteration complexity for the first-order optimization under an $L$-smooth condition and a bounded noise variance assumption. However, a thorough review of existing literature on Adam's convergence reveals a noticeable gap: none of them meet the above lower bound. In this paper, we close the gap by deriving a new convergence guarantee of Adam, with only an $L$-smooth condition and a bounded noise variance assumption. Our results remain valid across a broad spectrum of hyperparameters. Especially with properly chosen hyperparameters, we derive an upper bound of the iteration complexity of Adam and show that it meets the lower bound for first-order optimizers. To the best of our knowledge, this is the first to establish such a tight upper bound for Adam's convergence. Our proof utilizes novel techniques to handle the entanglement between momentum and adaptive learning rate and to convert the first-order term in the Descent Lemma to the gradient norm, which may be of independent interest.
摘要
近些时候，Arjevani等人（1）已经建立了first-order优化的迭代复杂度下界。然而，现有文献中对Adam的减少报告没有满足上述下界。在这篇论文中，我们填补了这一漏洞，通过引入$L$-smooth条件和bounded noise variance假设， derivate一个新的Adam的减少保证。我们的结论适用于广泛的权重参数。特别是，对于适当的权重参数，我们 derive一个迭代复杂度的 Upper bound of Adam，并证明它与first-order优化器的下界相符。根据我们知道，这是first-order优化器的减少保证的首次建立。我们的证明使用了新的技术来处理杠杆和自适应学习率的杂糅，并将 Descent Lemma 中的first-order项转换为梯度norm，这可能有独立的价值。
results: 这篇论文显示了一个叫做“碳能efficient FL”，其中使用了能源的碳气况来衡量成本。结果显示，这种方法可以 reduces carbon emissions by 93% 和 reduces training time by 50% 相比随机选择客户端。另外，它可以 reduces carbon emissions by 80%, 而仅增加训练时间 by 38% 相比一种现有的方法。Abstract
Federated Learning (FL) distributes machine learning (ML) training across many edge devices to reduce data transfer overhead and protect data privacy. Since FL model training may span millions of devices and is thus resource-intensive, prior work has focused on improving its resource efficiency to optimize time-to-accuracy. However, prior work generally treats all resources the same, while, in practice, they may incur widely different costs, which instead motivates optimizing cost-to-accuracy. To address the problem, we design CEFL, which uses adaptive cost-aware client selection policies to optimize an arbitrary cost metric when training FL models. Our policies extend and combine prior work on utility-based client selection and critical learning periods by making them cost-aware. We demonstrate CEFL by designing carbon-efficient FL, where energy's carbon-intensity is the cost, and show that it i) reduces carbon emissions by 93\% and reduces training time by 50% compared to random client selection and ii) reduces carbon emissions by 80%, while only increasing training time by 38%, compared to a state-of-the-art approach that optimizes training time.
摘要
协同学习（FL）通过分布机器学习训练 across多个边缘设备来减少数据传输开销和保护数据隐私。由于FL模型训练可能涵盖数百万个设备，因此需要进行资源效率优化以提高时间精度。然而，先前的工作通常忽视不同资源之间的差异，而在实践中，这些资源可能具有不同的成本，这些成本反而需要优化成本精度。为解决这个问题，我们提出了CEFL，它使用适应成本 aware的客户端选择策略来优化任意成本度量在协同学习模型训练中。我们的策略扩展和结合了先前的实用性基于资源利用率的客户端选择策略和批处理学习时期的优化策略，使其成为成本 aware。我们通过设计碳素协同学习，其中能源的碳气强度作为成本，并证明了它可以：1. 降低碳排放量93%，降低训练时间50%比Random Client Selection。2. 降低碳排放量80%，仅提高训练时间38%比一种状态精通的方法。
results: 本文提出了一些解决方案来提高 EML 系统的可靠性，但也指出了一些研究挑战和未解决的问题。Abstract
The convergence of Edge Computing (EC) and Machine Learning (ML), known as Edge Machine Learning (EML), has become a highly regarded research area by utilizing distributed network resources to perform joint training and inference in a cooperative manner. However, EML faces various challenges due to resource constraints, heterogeneous network environments, and diverse service requirements of different applications, which together affect the trustworthiness of EML in the eyes of its stakeholders. This survey provides a comprehensive summary of definitions, attributes, frameworks, techniques, and solutions for trustworthy EML. Specifically, we first emphasize the importance of trustworthy EML within the context of Sixth-Generation (6G) networks. We then discuss the necessity of trustworthiness from the perspective of challenges encountered during deployment and real-world application scenarios. Subsequently, we provide a preliminary definition of trustworthy EML and explore its key attributes. Following this, we introduce fundamental frameworks and enabling technologies for trustworthy EML systems, and provide an in-depth literature review of the latest solutions to enhance trustworthiness of EML. Finally, we discuss corresponding research challenges and open issues.
摘要随着边缘计算（EC）和机器学习（ML）的融合，称为边缘机器学习（EML），已经成为了非常受到关注的研究领域，通过分布式网络资源进行共同训练和推理，以实现共同的目标。但是，EML受到了资源约束、多样网络环境和不同应用程序的服务需求等多种挑战，这些挑战共同影响了EML的可靠性，从而影响了它的投资者和用户的信任度。本文提供了Edge Machine Learning的全面概述，包括定义、特征、框架、技术和解决方案，以确保EML在6G网络中的可靠性。Here's the breakdown of the translation: - This tag indicates that the following text is a system-level translation, rather than a word-for-word translation.随着边缘计算（EC）和机器学习（ML）的融合 - This phrase translates to "With the convergence of edge computing and machine learning."称为边缘机器学习（EML） - This phrase translates to "known as edge machine learning."已经成为了非常受到关注的研究领域 - This phrase translates to "has already become a highly regarded research area."通过分布式网络资源进行共同训练和推理 - This phrase translates to "by utilizing distributed network resources to perform joint training and inference."以实现共同的目标 - This phrase translates to "to achieve common goals."但是 - This word translates to "but."EML受到了资源约束、多样网络环境和不同应用程序的服务需求等多种挑战 - This phrase translates to "EML faces various challenges due to resource constraints, heterogeneous network environments, and diverse service requirements of different applications."这些挑战共同影响了EML的可靠性 - This phrase translates to "these challenges collectively affect the trustworthiness of EML."从而影响了它的投资者和用户的信任度 - This phrase translates to "and thus affect the investors and users' trust in it."本文提供了Edge Machine Learning的全面概述，包括定义、特征、框架、技术和解决方案 - This phrase translates to "This article provides a comprehensive overview of Edge Machine Learning, including definitions, features, frameworks, techniques, and solutions."以确保EML在6G网络中的可靠性 - This phrase translates to "to ensure the reliability of EML in 6G networks."
MicroNAS: Memory and Latency Constrained Hardware-Aware Neural Architecture Search for Time Series Classification on Microcontrollers
results: 这个系统可以实现高精度的识别结果（UCI-HAR 93.93%，SkodaR 96.33%），并在 Cortex-M4 MCU 上运行。Abstract
This paper presents MicroNAS, a system designed to automatically search and generate neural network architectures capable of classifying time series data on resource-constrained microcontrollers (MCUs) and generating standard tf-lite ML models. MicroNAS takes into account user-defined constraints on execution latency and peak memory consumption on a target MCU. This approach ensures that the resulting neural network architectures are optimised for the specific constraints and requirements of the MCU on which they are implemented. To achieve this, MicroNAS uses a look-up table estimation approach for accurate execution latency calculations, with a minimum error of only 1.02ms. This accurate latency estimation on MCUs sets it apart from other hardware-aware neural architecture search (HW-NAS) methods that use less accurate estimation techniques. Finally, MicroNAS delivers performance close to that of state-of-the-art models running on desktop computers, achieving high classification accuracies on recognised datasets (93.93% on UCI-HAR and 96.33% on SkodaR) while running on a Cortex-M4 MCU.
摘要
To achieve this, MicroNAS uses a look-up table estimation approach for accurate execution latency calculations, with a minimum error of only 1.02 milliseconds. This accurate latency estimation on MCUs sets it apart from other hardware-aware neural architecture search (HW-NAS) methods that use less accurate estimation techniques.Finally, MicroNAS delivers performance close to that of state-of-the-art models running on desktop computers, achieving high classification accuracies on recognized datasets (93.93% on UCI-HAR and 96.33% on SkodaR) while running on a Cortex-M4 MCU.
Lifting the Veil: Unlocking the Power of Depth in Q-learning
paper_authors: Shao-Bo Lin, Tao Li, Shaojie Tang, Yao Wang, Ding-Xuan Zhou
for: 本文试图 theoretically verify the power of depth in deep Q-learning, and provide a solid theoretical foundation for its success in numerous applications.
methods: 本文使用 statistical learning theory 来rigorously prove that deep Q-learning outperforms traditional Q-learning, and demonstrate its good generalization error bound.
results: 研究发现，深度 Q-learning 的成功主要归功于深度神经网络（deep nets）的特殊性能，而不是它的大容量。 Additionally, the paper answers three key questions: Why does deep Q-learning perform so well? When does deep Q-learning perform better than traditional Q-learning? How many samples are required to achieve a specific prediction accuracy for deep Q-learning?Abstract
With the help of massive data and rich computational resources, deep Q-learning has been widely used in operations research and management science and has contributed to great success in numerous applications, including recommender systems, supply chains, games, and robotic manipulation. However, the success of deep Q-learning lacks solid theoretical verification and interpretability. The aim of this paper is to theoretically verify the power of depth in deep Q-learning. Within the framework of statistical learning theory, we rigorously prove that deep Q-learning outperforms its traditional version by demonstrating its good generalization error bound. Our results reveal that the main reason for the success of deep Q-learning is the excellent performance of deep neural networks (deep nets) in capturing the special properties of rewards namely, spatial sparseness and piecewise constancy, rather than their large capacities. In this paper, we make fundamental contributions to the field of reinforcement learning by answering to the following three questions: Why does deep Q-learning perform so well? When does deep Q-learning perform better than traditional Q-learning? How many samples are required to achieve a specific prediction accuracy for deep Q-learning? Our theoretical assertions are verified by applying deep Q-learning in the well-known beer game in supply chain management and a simulated recommender system.
摘要
通过庞大的数据和丰富的计算资源，深度Q学习在运筐研究和管理科学中广泛应用，并在多个应用中做出了卓越的成绩，包括推荐系统、供应链、游戏和机器人操作。然而，深度Q学习的成功尚未得到坚实的理论验证和可读性。本文的目标是从统计学学习理论的视角出发，确认深度Q学习的力量。我们在统计学学习理论的框架下，严格地证明了深度Q学习的泛化误差 bound 比传统Q学习更好。我们的结果表明，深度Q学习的成功主要归功于深度神经网络（深度网）在奖励特性上表现出色，而不是它的大容量。本文对抗习学习领域做出了基础性的贡献，回答了以下三个问题：深度Q学习为什么会表现 så well? 深度Q学习在哪些情况下表现更好于传统Q学习? 深度Q学习需要多少样本来达到特定的预测精度？我们的理论声明得到了在啤酒游戏和一个模拟的推荐系统中应用深度Q学习的实质验证。
results: 作者提出了一种改进的策略，即改进知识梯度（iKG）策略，它可以在 variant problems of BAI 中展现出更好的性能。在数学示例中，iKG 的性能也被证明是比 KG 更好的。Abstract
The knowledge gradient (KG) algorithm is a popular policy for the best arm identification (BAI) problem. It is built on the simple idea of always choosing the measurement that yields the greatest expected one-step improvement in the estimate of the best mean of the arms. In this research, we show that this policy has limitations, causing the algorithm not asymptotically optimal. We next provide a remedy for it, by following the manner of one-step look ahead of KG, but instead choosing the measurement that yields the greatest one-step improvement in the probability of selecting the best arm. The new policy is called improved knowledge gradient (iKG). iKG can be shown to be asymptotically optimal. In addition, we show that compared to KG, it is easier to extend iKG to variant problems of BAI, with the $\epsilon$-good arm identification and feasible arm identification as two examples. The superior performances of iKG on these problems are further demonstrated using numerical examples.
摘要
“知识梯度（KG）算法是一种受欢迎的策略 для最佳臂 Identification（BAI）问题。它基于简单的想法，就是总是选择测量，可以将最大化预期的一步改善在臂的最佳均值的估计。在这个研究中，我们显示出这个策略有限制，导致算法不是 asymptotically 优化的。我们随后提供了一个修正方案，通过一步前进的方式，选择测量，可以将最大化一步改善在臂选择的可能性。这个新策略被称为改善知识梯度（iKG）。iKG可以显示是 asymptotically 优化的。此外，我们显示了在 variant 问题中，iKG比KG更容易扩展，例如 $\epsilon$-good arm identification 和可行臂 identification 两个例子。iKG 在这些问题上的表现更加出色，通过数学例子进行说明。”Note: The translation is in Simplified Chinese, which is the standard form of Chinese used in mainland China and Singapore. If you need Traditional Chinese, please let me know.
Submodel Partitioning in Hierarchical Federated Learning: Algorithm Design and Convergence Analysis
results: 通过数学分析和实验 validate HIST 能够在非对称损失函数下保证收敛性，并且在许多属性（如Cell数量、本地和全球聚合频率）的影响下对性能与效率进行了评估。实验结果表明，HIST 能够大幅减少通信成本，同时保持测试准确率不变。Abstract
Hierarchical federated learning (HFL) has demonstrated promising scalability advantages over the traditional "star-topology" architecture-based federated learning (FL). However, HFL still imposes significant computation, communication, and storage burdens on the edge, especially when training a large-scale model over resource-constrained Internet of Things (IoT) devices. In this paper, we propose hierarchical independent submodel training (HIST), a new FL methodology that aims to address these issues in hierarchical settings. The key idea behind HIST is a hierarchical version of model partitioning, where we partition the global model into disjoint submodels in each round, and distribute them across different cells, so that each cell is responsible for training only one partition of the full model. This enables each client to save computation/storage costs while alleviating the communication loads throughout the hierarchy. We characterize the convergence behavior of HIST for non-convex loss functions under mild assumptions, showing the impact of several attributes (e.g., number of cells, local and global aggregation frequency) on the performance-efficiency tradeoff. Finally, through numerical experiments, we verify that HIST is able to save communication costs by a wide margin while achieving the same target testing accuracy.
摘要
HIST 的关键想法是在每轮中对全球模型进行层次分区，将每个分区分配给不同的细胞，以便每个客户端只需训练自己的分区，而不需要与其他细胞进行通信。这样，每个客户端都可以降低计算/存储成本，同时减轻通信负担。我们分析了不同参数（例如细胞数、本地和全球汇总频率）对性能与效率的影响。最后，通过数值实验，我们证明了 HIST 可以减少通信成本，同时保持测试准确率不变。
Machine Learning Infused Distributed Optimization for Coordinating Virtual Power Plant Assets
paper_authors: Meiyi Li, Javad Mohammadi for: This paper aims to present a novel machine learning-assisted distributed optimization method for coordinating Virtual Power Plants (VPPs) and their associated Distributed Energy Resources (DERs).methods: The proposed method, named LOOP-MAC, utilizes a multi-agent coordination approach and neural network approximators to expedite the solution search.results: The LOOP-MAC method demonstrates accelerated solution times per iteration and significantly reduced convergence times compared to conventional centralized and distributed optimization methods.Abstract
Amid the increasing interest in the deployment of Distributed Energy Resources (DERs), the Virtual Power Plant (VPP) has emerged as a pivotal tool for aggregating diverse DERs and facilitating their participation in wholesale energy markets. These VPP deployments have been fueled by the Federal Energy Regulatory Commission's Order 2222, which makes DERs and VPPs competitive across market segments. However, the diversity and decentralized nature of DERs present significant challenges to the scalable coordination of VPP assets. To address efficiency and speed bottlenecks, this paper presents a novel machine learning-assisted distributed optimization to coordinate VPP assets. Our method, named LOOP-MAC(Learning to Optimize the Optimization Process for Multi-agent Coordination), adopts a multi-agent coordination perspective where each VPP agent manages multiple DERs and utilizes neural network approximators to expedite the solution search. The LOOP-MAC method employs a gauge map to guarantee strict compliance with local constraints, effectively reducing the need for additional post-processing steps. Our results highlight the advantages of LOOP-MAC, showcasing accelerated solution times per iteration and significantly reduced convergence times. The LOOP-MAC method outperforms conventional centralized and distributed optimization methods in optimization tasks that require repetitive and sequential execution.
摘要
在增加分布能源资源（DERs）的投入中，虚拟发电厂（VPP）已成为汇集多种DERs并促进其参与到总体能源市场中的关键工具。这些VPP部署受到联邦能源管理委员会的命令2222的推动，该命令使DERs和VPPs在市场 segments中竞争。然而，DERs的多样性和分散化带来了VPP资产的扩展协调的显著挑战。为了提高效率和速度瓶颈，本文提出了一种新的机器学习协助分布优化方法，称为LOOP-MAC（学习优化优化过程多代理协调）。LOOP-MAC方法采用多代理协调视角，每个VPP代理负责多个DERs，并使用神经网络approximators快速搜索解决方案。LOOP-MAC方法使用一个报表图来保证本地约束的严格遵从，从而减少了额外处理步骤的需要。我们的结果表明LOOP-MAC方法具有优势，其解决时间和趋势时间均显著减少。LOOP-MAC方法在需要重复和序列执行的优化任务中超过了传统中央化和分布式优化方法。
A Sublinear-Time Spectral Clustering Oracle with Improved Preprocessing Time
results: 本文的结果表明，对于具有较大内径导通率（至少为 $\varphi$）和较小外径导通率（至多为 $\varepsilon$）的图像，可以实现高效的 clustering membership queries。Abstract
We address the problem of designing a sublinear-time spectral clustering oracle for graphs that exhibit strong clusterability. Such graphs contain $k$ latent clusters, each characterized by a large inner conductance (at least $\varphi$) and a small outer conductance (at most $\varepsilon$). Our aim is to preprocess the graph to enable clustering membership queries, with the key requirement that both preprocessing and query answering should be performed in sublinear time, and the resulting partition should be consistent with a $k$-partition that is close to the ground-truth clustering. Previous oracles have relied on either a $\textrm{poly}(k)\log n$ gap between inner and outer conductances or exponential (in $k/\varepsilon$) preprocessing time. Our algorithm relaxes these assumptions, albeit at the cost of a slightly higher misclassification ratio. We also show that our clustering oracle is robust against a few random edge deletions. To validate our theoretical bounds, we conducted experiments on synthetic networks.
摘要
我们处理拥有强 clustering 性的图的问题，特别是对于具有至少 $\phi$ 内导通和最多 $\varepsilon$ 外导通的 $k$ 个粒子集的图。我们的目标是在子线性时间内进行类别查询，并且保持类别结果与真实分类相似。previous 的标志有两个假设：一是 $\textrm{poly}(k)\log n$ 内外导通之间的差，二是 exponential （在 $k/\varepsilon$ 上）的预processing 时间。我们的算法则关注这两个假设，但是会导致轻微的错分率增加。我们还证明了我们的类别标志对于几个随机边的删除而Robust。为了证明我们的理论上限，我们对Synthetic 网络进行实验。
From Generative AI to Generative Internet of Things: Fundamentals, Framework, and Outlooks
results: 文章通过一个现有的现场案例研究，利用生成扩散模型（GDM）设计有效的奖励合同，以吸引用户提供高质量感知数据。此外，文章还提出了一些未来研究的开放方向。Abstract
Generative Artificial Intelligence (GAI) possesses the capabilities of generating realistic data and facilitating advanced decision-making. By integrating GAI into modern Internet of Things (IoT), Generative Internet of Things (GIoT) is emerging and holds immense potential to revolutionize various aspects of society, enabling more efficient and intelligent IoT applications, such as smart surveillance and voice assistants. In this article, we present the concept of GIoT and conduct an exploration of its potential prospects. Specifically, we first overview four GAI techniques and investigate promising GIoT applications. Then, we elaborate on the main challenges in enabling GIoT and propose a general GAI-based secure incentive mechanism framework to address them, in which we adopt Generative Diffusion Models (GDMs) for incentive mechanism designs and apply blockchain technologies for secure GIoT management. Moreover, we conduct a case study on modern Internet of Vehicle traffic monitoring, which utilizes GDMs to generate effective contracts for incentivizing users to contribute sensing data with high quality. Finally, we suggest several open directions worth investigating for the future popularity of GIoT.
摘要
优化的人工智能（GAI）具有生成真实数据和提供高级决策的能力。将GAI融入现代互联网器（IoT）后，生成互联网器（GIoT）得到了巨大的潜力，推动了社会各方面的改革，如智能监测和语音助手等智能应用。本文提出了GIoT的概念，并对其潜在可能性进行了探讨。 Specifically，我们首先介绍了四种GAI技术，然后研究了GIoT的潜在应用场景。然后，我们详细介绍了GIoT实现的主要挑战和一种基于GAI的安全奖励机制框架，其中采用生成扩散模型（GDMs）为奖励机制设计，并应用区块链技术来安全地管理GIoT。此外，我们进行了现代互联网器交通监测的实践案例，利用GDMs生成高质量感知数据的合法合约。最后，我们提出了未来GIoT的一些开放方向值得进一步探索。
Unveil Sleep Spindles with Concentration of Frequency and Time
For: The paper aims to develop an accurate and interpretable algorithm for sleep spindle detection in EEG data, and to quantify the instantaneous frequencies of spindles.* Methods: The authors introduce a novel non-linear time-frequency analysis tool called “Concentration of Frequency and Time” (ConceFT), which effectively reduces stochastic EEG influence and enhances spindle visibility in the time-frequency representation. They also developed an automated spindle detection algorithm called ConceFT-Spindle (ConceFT-S), which is compared to two other algorithms (A7 and SUMO) using two benchmark databases (Dream and MASS).* Results: The results show that ConceFT-S achieves F1 scores of 0.749 in Dream and 0.786 in MASS, which is equivalent to or surpasses the performance of A7 and SUMO with statistical significance. Additionally, the authors reveal that spindle IF is generally nonlinear.Here are the three points in Simplified Chinese text:* For: 这个论文目的是开发一个准确和可解释的EEG数据中睡眠潮汐检测算法，并量化潮汐的快速频率。* Methods: 作者们引入了一种新的非线性时间频谱分析工具”集中频率和时间”(ConceFT),该工具有效减少了随机EEG的影响，使潮汐在时间频谱表示中更加明了潮汐。他们还开发了一个自动潮汐检测算法ConceFT-Spindle(ConceFT-S)，并与A7和SUMO两个算法进行比较使用了两个标准数据库(Dream和MASS)。* Results: 结果表明，ConceFT-S在Dream和MASS两个数据库中的F1分数分别为0.749和0.786，这与或超过A7和SUMO的性能有统计学上的显著性。此外，作者们还发现，潮汐的快速频率通常是非线性的。Abstract
Objective: Sleep spindles contain crucial brain dynamics information. We introduce the novel non-linear time-frequency analysis tool 'Concentration of Frequency and Time' (ConceFT) to create an interpretable automated algorithm for sleep spindle annotation in EEG data and to measure spindle instantaneous frequencies (IFs). Methods: ConceFT effectively reduces stochastic EEG influence, enhancing spindle visibility in the time-frequency representation. Our automated spindle detection algorithm, ConceFT-Spindle (ConceFT-S), is compared to A7 (non-deep learning) and SUMO (deep learning) using Dream and MASS benchmark databases. We also quantify spindle IF dynamics. Results: ConceFT-S achieves F1 scores of 0.749 in Dream and 0.786 in MASS, which is equivalent to or surpass A7 and SUMO with statistical significance. We reveal that spindle IF is generally nonlinear. Conclusion: ConceFT offers an accurate, interpretable EEG-based sleep spindle detection algorithm and enables spindle IF quantification.
摘要
目标：睡眠尖峰含有关键脑动态信息。我们介绍了一种新的非线性时域分析工具“时域频率卷积”（ConceFT），以创建可解释的自动化睡眠尖峰标注算法，并测量尖峰快速频率（IF）的动态变化。方法：ConceFT可以有效减少随机的EEG影响，使睡眠尖峰在时域表示更加明显。我们的自动化睡眠尖峰检测算法ConceFT-Spindle（ConceFT-S）与A7（非深度学习）和SUMO（深度学习）在梦境和MASS数据库上进行比较，并评估尖峰IF动态变化。结果：ConceFT-S在梦境和MASS上的F1分数分别为0.749和0.786，与A7和SUMO相当或超过，这种差异为统计学上的显著性。我们发现尖峰IF通常是非线性的。结论：ConceFT提供了一种准确可解释的EEG基于睡眠尖峰检测算法，并允许量化尖峰IF的动态变化。
Boosting Data Analytics With Synthetic Volume Expansion
results: 研究发现，随着生成数据的增加，统计方法的误差最初逐渐减少，但 eventually可能增加或折衣。这种现象被称为“生成效应”，它表明在生成数据中复制原始数据的分布时存在一个“反射点”，即特定的误差度量的优化阈值。通过三个案例研究，包括文本感知分析、结构化数据预测和表格数据推理，我们证明了这种框架的效果，并将其与传统方法进行比较。Abstract
Synthetic data generation, a cornerstone of Generative Artificial Intelligence, signifies a paradigm shift in data science by addressing data scarcity and privacy while enabling unprecedented performance. As synthetic data gains prominence, questions arise concerning the accuracy of statistical methods when applied to synthetic data compared to raw data. In this article, we introduce the Synthetic Data Generation for Analytics framework. This framework employs statistical methods on high-fidelity synthetic data generated by advanced models such as tabular diffusion and Generative Pre-trained Transformer models. These models, trained on raw data, are further enhanced with insights from pertinent studies. A significant discovery within this framework is the generational effect: the error of a statistical method on synthetic data initially diminishes with added synthetic data but may eventually increase or plateau. This phenomenon, rooted in the complexities of replicating raw data distributions, highlights a "reflection point"--an optimal threshold in the size of synthetic data determined by specific error metrics. Through three illustrative case studies-sentiment analysis of texts, predictive modeling of structured data, and inference in tabular data--we demonstrate the effectiveness of this framework over traditional ones. We underline its potential to amplify various statistical methods, including gradient boosting for prediction and hypothesis testing, thereby underscoring the transformative potential of synthetic data generation in data science.
摘要
<>translate "Synthetic data generation, a cornerstone of Generative Artificial Intelligence, signifies a paradigm shift in data science by addressing data scarcity and privacy while enabling unprecedented performance. As synthetic data gains prominence, questions arise concerning the accuracy of statistical methods when applied to synthetic data compared to raw data. In this article, we introduce the Synthetic Data Generation for Analytics framework. This framework employs statistical methods on high-fidelity synthetic data generated by advanced models such as tabular diffusion and Generative Pre-trained Transformer models. These models, trained on raw data, are further enhanced with insights from pertinent studies. A significant discovery within this framework is the generational effect: the error of a statistical method on synthetic data initially diminishes with added synthetic data but may eventually increase or plateau. This phenomenon, rooted in the complexities of replicating raw data distributions, highlights a "reflection point"--an optimal threshold in the size of synthetic data determined by specific error metrics. Through three illustrative case studies-sentiment analysis of texts, predictive modeling of structured data, and inference in tabular data--we demonstrate the effectiveness of this framework over traditional ones. We underline its potential to amplify various statistical methods, including gradient boosting for prediction and hypothesis testing, thereby underscoring the transformative potential of synthetic data generation in data science."中文简体版：<>生成数据领域，人工智能生成的核心，数据科学领域发生了一场 парадигShift，通过地址数据缺乏和隐私问题，实现了无前例的性能。随着生成数据的普及，关注统计方法在生成数据上的准确性问题 arise。本文介绍了生成数据分析框架。这个框架利用高准确度的生成数据，由高级模型如表 diffusion和生成预训练 transformer 模型生成，这些模型在原始数据上训练。在这个框架中，我们发现了一种生成效应：在生成数据上使用统计方法的错误在初始阶段随着添加生成数据减少，但可能在某些点上增加或稳定。这种现象基于生成数据 Distribution 复杂性，表明一个 "反射点"--一个特定的错误指标决定的最佳大小。通过三个案例研究--文本情感分析、结构化数据预测和表格数据推理--我们示出了这个框架的效果，比传统方法更高。我们强调了它的潜在作用，包括权度提升、预测和假设测试，从而强调生成数据生成在数据科学中的转型潜力。
A Data-Centric Online Market for Machine Learning: From Discovery to Pricing
results: 论文的实验结果表明，这些新技术可以有效地匹配ML任务和数据，并且可以鼓励ML用户参与到市场中，从而提高ML模型的性能和可用性。Abstract
Data fuels machine learning (ML) - rich and high-quality training data is essential to the success of ML. However, to transform ML from the race among a few large corporations to an accessible technology that serves numerous normal users' data analysis requests, there still exist important challenges. One gap we observed is that many ML users can benefit from new data that other data owners possess, whereas these data owners sit on piles of data without knowing who can benefit from it. This gap creates the opportunity for building an online market that can automatically connect supply with demand. While online matching markets are prevalent (e.g., ride-hailing systems), designing a data-centric market for ML exhibits many unprecedented challenges. This paper develops new techniques to tackle two core challenges in designing such a market: (a) to efficiently match demand with supply, we design an algorithm to automatically discover useful data for any ML task from a pool of thousands of datasets, achieving high-quality matching between ML models and data; (b) to encourage market participation of ML users without much ML expertise, we design a new pricing mechanism for selling data-augmented ML models. Furthermore, our market is designed to be API-compatible with existing online ML markets like Vertex AI and Sagemaker, making it easy to use while providing better results due to joint data and model search. We envision that the synergy of our data and model discovery algorithm and pricing mechanism will be an important step towards building a new data-centric online market that serves ML users effectively.
摘要
数据驱动机器学习（ML）——高质量和丰富的训练数据是ML的成功关键。然而，将ML从几家大公司的竞赛转变为让常见用户的数据分析请求的可 accessible 技术，仍存在重要挑战。我们发现了一个差距：许多ML用户可以从其他数据所有者手中获得新的数据，而这些数据所有者拥有大量数据，不知道谁可以从中受益。这个差距创造了建立一个在线市场的机会，可以自动连接供应和需求。虽然在线匹配市场是普遍的（例如，乘车应用程序），但设计一个专门为ML的数据市场具有许多前所未有的挑战。本文提出了新的技术来解决两个核心挑战：(a) 高效匹配需求和供应，我们设计了一个自动从千余个数据集中找到适用于任何ML任务的有用数据，以实现高质量的匹配 междуML模型和数据。(b) 鼓励ML用户 без ML专业知识参与市场，我们设计了一种新的价格机制来销售数据增强ML模型。此外，我们的市场采用API兼容于现有的在线ML市场 like Vertex AI和Sagemaker，使其易于使用，同时提供更好的结果由于共同数据和模型搜索。我们认为，我们的数据和模型发现算法和价格机制的共同作用将是建立一个新的数据驱动的在线市场的重要一步。
Positional Encoding-based Resident Identification in Multi-resident Smart Homes
results: 广泛的实验表明，提出的方案可以有效地识别多名occupant在智能环境中。两个实际数据集的评估结果显示，该方案的准确率分别为94.5%和87.9%。Abstract
We propose a novel resident identification framework to identify residents in a multi-occupant smart environment. The proposed framework employs a feature extraction model based on the concepts of positional encoding. The feature extraction model considers the locations of homes as a graph. We design a novel algorithm to build such graphs from layout maps of smart environments. The Node2Vec algorithm is used to transform the graph into high-dimensional node embeddings. A Long Short-Term Memory (LSTM) model is introduced to predict the identities of residents using temporal sequences of sensor events with the node embeddings. Extensive experiments show that our proposed scheme effectively identifies residents in a multi-occupant environment. Evaluation results on two real-world datasets demonstrate that our proposed approach achieves 94.5% and 87.9% accuracy, respectively.
摘要
我们提出了一种新的居民标识框架，用于在多occupant智能环境中识别居民。我们的框架使用基于 pozitional 编码的特征提取模型，该模型考虑了智能环境的布局地图。我们提出了一种新的算法，用于从布局地图中生成图形。然后，我们使用 Node2Vec 算法将图形转换成高维节点嵌入。我们引入了一个长期快速储存（LSTM）模型，用于预测基于时间序列的感知事件的居民身份。广泛的实验表明，我们的提议方案可以有效地识别多occupant环境中的居民。两个实际数据集的评估结果表明，我们的方法可以达到94.5%和87.9%的准确率。
Hybrid Optical Turbulence Models Using Machine Learning and Local Measurements
results: 研究发现，这个混合模型可以比基本的大气macro-meteorological模型和机器学习模型更好地预测大气光学震荡的性能，尤其是在训练数据少的情况下。Abstract
Accurate prediction of atmospheric optical turbulence in localized environments is essential for estimating the performance of free-space optical systems. Macro-meteorological models developed to predict turbulent effects in one environment may fail when applied in new environments. However, existing macro-meteorological models are expected to offer some predictive power. Building a new model from locally-measured macro-meteorology and scintillometer readings can require significant time and resources, as well as a large number of observations. These challenges motivate the development of a machine-learning informed hybrid model framework. By combining some baseline macro-meteorological model with local observations, hybrid models were trained to improve upon the predictive power of each baseline model. Comparisons between the performance of the hybrid models, the selected baseline macro-meteorological models, and machine-learning models trained only on local observations highlight potential use cases for the hybrid model framework when local data is expensive to collect. Both the hybrid and data-only models were trained using the Gradient Boosted Decision Tree (GBDT) architecture with a variable number of in-situ meteorological observations. The hybrid and data-only models were found to outperform three baseline macro-meteorological models, even for low numbers of observations, in some cases as little as one day. For the first baseline macro-meteorological model investigated, the hybrid model achieves an estimated 29% reduction in mean absolute error (MAE) using only one days-equivalent of observation, growing to 41% after only two days, and 68% after 180 days-equivalent training data. The number of days-equivalent training data required is potentially indicative of the seasonal variation in the local microclimate and its propagation environment.
摘要
准确预测大气光学抖振在本地化环境中是自由空间光学系统性能预测的关键。 macro-метеорологические模型在不同环境中预测抖振效果可能失败，但现有的 macro-метеорологические模型仍然可以提供一定的预测力。 基于本地测量的 macro-метеорологи和抖振仪读数建立新模型可能需要较长的时间和资源，以及大量观测数据。这些挑战驱动了开发一种机器学习 Informed 混合模型框架。通过将基线 macro-метеорологических模型与本地观测数据结合，混合模型可以提高每个基线模型的预测力。对比 hybrid 模型、选择的基线 macro-метеорологи models 和只使用本地观测数据训练的机器学习模型， hybrid 模型在一些情况下可以更好地预测抖振效果。使用 Gradient Boosted Decision Tree (GBDT) 架构，hybrid 模型和数据 только模型都被训练使用本地 meteorological 观测数据。在一些情况下，hybrid 模型可以在只需一天的观测数据量下表现出较好的预测效果，而且随着训练数据量的增加，hybrid 模型的性能会得到进一步改善。对于第一个基eline macro-метеорологи models investigated，hybrid 模型可以在一天的训练数据量下实现了相对于基线模型的29%的减少 Mean Absolute Error (MAE)，随着训练数据量的增加，hybrid 模型的性能会得到进一步改善。这些结果表明了hybrid 模型在本地数据质量较低的情况下的可行性。 hybrid 模型和数据只模型的训练需要的天数相对于季节变化的本地微气候和其传播环境可能有关。