results: 本文的分析表明，VoX 算法可以在低维 MDPs 中提供 provably 样本效率的探索，并且不需要额外的模型基础或 latent variable structure。Abstract
A major challenge in reinforcement learning is to develop practical, sample-efficient algorithms for exploration in high-dimensional domains where generalization and function approximation is required. Low-Rank Markov Decision Processes -- where transition probabilities admit a low-rank factorization based on an unknown feature embedding -- offer a simple, yet expressive framework for RL with function approximation, but existing algorithms are either (1) computationally intractable, or (2) reliant upon restrictive statistical assumptions such as latent variable structure, access to model-based function approximation, or reachability. In this work, we propose the first provably sample-efficient algorithm for exploration in Low-Rank MDPs that is both computationally efficient and model-free, allowing for general function approximation and requiring no additional structural assumptions. Our algorithm, VoX, uses the notion of a generalized optimal design for the feature embedding as an efficiently computable basis for exploration, performing efficient optimal design computation by interleaving representation learning and policy optimization. Our analysis -- which is appealingly simple and modular -- carefully combines several techniques, including a new reduction from optimal design computation to policy optimization based on the Frank-Wolfe method, and an improved analysis of a certain minimax representation learning objective found in prior work.
摘要
一个主要挑战在强化学习中是开发实用、样本效率高的探索算法，以便在高维度空间中进行探索，并且需要泛化和函数近似。低维马尔可夫遇过程（Low-Rank MDPs）提供了一个简单 yet 表达力强的框架，但现有算法的问题包括（1）计算复杂度太高，或（2）需要特殊的统计假设，如隐藏变量结构、函数近似模型或可达性。在这项工作中，我们提出了首个可证明样本效率高的探索算法，可以在低维马尔可夫遇过程中进行探索，并且不需要特殊的结构假设。我们的算法VoX利用了一个通用优化设计的概念，作为一种可读取的基础 для探索，并通过将表示学习和政策优化结合在一起，实现高效的优化设计计算。我们的分析，感知简单而干净，综合使用了多种技术，包括一种新的减少从优化设计计算到政策优化基于Frank-Wolfe方法的技术，以及在先前的工作中发现的一种改进的最小最大表达学习目标的分析。
NLP Meets RNA: Unsupervised Embedding Learning for Ribozymes with Word2Vec
results: 结果表明，使用核酸序列embedding可以准确地分类核酸杂合物，并且256维度的embedding Vector Space可以捕捉核酸杂合物的特征。Abstract
Ribozymes, RNA molecules with distinct 3D structures and catalytic activity, have widespread applications in synthetic biology and therapeutics. However, relatively little research has focused on leveraging deep learning to enhance our understanding of ribozymes. This study implements Word2Vec, an unsupervised learning technique for natural language processing, to learn ribozyme embeddings. Ribo2Vec was trained on over 9,000 diverse ribozymes, learning to map sequences to 128 and 256-dimensional vector spaces. Using Ribo2Vec, sequence embeddings for five classes of ribozymes (hatchet, pistol, hairpin, hovlinc, and twister sister) were calculated. Principal component analysis demonstrated the ability of these embeddings to distinguish between ribozyme classes. Furthermore, a simple SVM classifier trained on ribozyme embeddings showed promising results in accurately classifying ribozyme types. Our results suggest that the embedding vectors contained meaningful information about ribozymes. Interestingly, 256-dimensional embeddings behaved similarly to 128-dimensional embeddings, suggesting that a lower dimension vector space is generally sufficient to capture ribozyme features. This approach demonstrates the potential of Word2Vec for bioinformatics, opening new avenues for ribozyme research. Future research includes using a Transformer-based method to learn RNA embeddings, which can capture long-range interactions between nucleotides.
摘要
瑞博酵素（ribozyme）具有广泛的应用前景，包括生物技术和生物医学。然而，相对论文不多关注利用深度学习来提高我们对瑞博酵素的理解。这项研究使用Word2Vec算法，一种无监督学习技术，来学习瑞博酵素的嵌入。 Ribo2Vec 训练在超过9000个多样化瑞博酵素上，将序列映射到128和256维度的向量空间中。使用 Ribo2Vec，我们计算了5种瑞博酵素类型（锥子、手枪、捷径、托征和姐妹）的序列嵌入。对于这些嵌入，我们使用主成分分析得到了可以分辨瑞博酵素类型的结果。此外，使用瑞博酵素嵌入训练的简单支持向量机器学习（SVM）分类器也表现出了良好的准确率。这些结果表明瑞博酵素嵌入 vectors 含有有用的信息。有趣的是，256维度的嵌入和128维度的嵌入之间的行为相似，这表明低维度向量空间通常足够 capture瑞博酵素特征。这种方法可能会打开新的 Bioinformatics 研究途径。未来的研究可能包括使用Transformer算法来学习RNA嵌入，以捕捉RNA中距离较远的核苷酸之间的长距离交互。
Integrating Curricula with Replays: Its Effects on Continual Learning
results: 我们的结果表明，这三种curricula有效地遏制了衰弱性失忆，并提高了正面知识传递。这些结果表明，curricula可以在持续学习方法中提供进一步的改进。我们的代码和数据可以在GitHub上找到：https://github.com/ZhangLab-DeepNeuroCogLab/Integrating-Curricula-with-ReplaysAbstract
Humans engage in learning and reviewing processes with curricula when acquiring new skills or knowledge. This human learning behavior has inspired the integration of curricula with replay methods in continual learning agents. The goal is to emulate the human learning process, thereby improving knowledge retention and facilitating learning transfer. Existing replay methods in continual learning agents involve the random selection and ordering of data from previous tasks, which has shown to be effective. However, limited research has explored the integration of different curricula with replay methods to enhance continual learning. Our study takes initial steps in examining the impact of integrating curricula with replay methods on continual learning in three specific aspects: the interleaved frequency of replayed exemplars with training data, the sequence in which exemplars are replayed, and the strategy for selecting exemplars into the replay buffer. These aspects of curricula design align with cognitive psychology principles and leverage the benefits of interleaved practice during replays, easy-to-hard rehearsal, and exemplar selection strategy involving exemplars from a uniform distribution of difficulties. Based on our results, these three curricula effectively mitigated catastrophic forgetting and enhanced positive knowledge transfer, demonstrating the potential of curricula in advancing continual learning methodologies. Our code and data are available: https://github.com/ZhangLab-DeepNeuroCogLab/Integrating-Curricula-with-Replays
摘要
人类在学习和复习过程中使用课程，以获得新技能或知识。这种人类学习行为对于 continual learning agent 的 интеграцию有着灵感。目标是通过模拟人类学习过程，提高知识保持和学习传递。现有的重播方法在 continual learning agent 中已经证明有效。然而，有限的研究探讨了不同课程的 интеграция和重播方法的影响。我们的研究首先探讨了在三个方面中的课程设计对 continual learning 的影响：重播示例的频率与训练数据的排序、示例的重播顺序和选择示例进入缓存的策略。这些课程设计方面与认知心理学原理相吻合，利用重播中的叠加实践、易于困难的复习和选择示例的策略。根据我们的结果，这三种课程有效地遏止了恶性遗忘和提高了正面知识传递，这表明了课程在 continual learning 方法中的潜力。我们的代码和数据可以在 GitHub 上找到：https://github.com/ZhangLab-DeepNeuroCogLab/Integrating-Curricula-with-Replays
Building and Road Segmentation Using EffUNet and Transfer Learning Approach
methods: 该论文提出了一种基于 Google 新提出的 EfficientNetV2 增强网络，并结合 UNet 解码器实现分割图像的方法。
results: 该方法在麻省建筑和路径数据集上达到了 benchmark 分割精度，IOU 分割精度分别为 0.8365 和 0.9153。Abstract
In city, information about urban objects such as water supply, railway lines, power lines, buildings, roads, etc., is necessary for city planning. In particular, information about the spread of these objects, locations and capacity is needed for the policymakers to make impactful decisions. This thesis aims to segment the building and roads from the aerial image captured by the satellites and UAVs. Many different architectures have been proposed for the semantic segmentation task and UNet being one of them. In this thesis, we propose a novel architecture based on Google's newly proposed EfficientNetV2 as an encoder for feature extraction with UNet decoder for constructing the segmentation map. Using this approach we achieved a benchmark score for the Massachusetts Building and Road dataset with an mIOU of 0.8365 and 0.9153 respectively.
摘要
在城市中，有关城市 объекts 如水 supply、铁路线、电力线、建筑、路径等的信息是城市规划的必要条件。特别是政策制定者需要知道这些对象的扩散、位置和容量，以便做出有力的决策。本论文目的是从航天图像和无人机拍摄的卫星和无人机图像中分割建筑和路径。许多不同的架构已经为semantic segmentation任务提出了多种方案，其中UNet是其中之一。本论文提出了基于Google新提出的EfficientNetV2架构来进行特征提取，并与UNet决码器结合使用以生成分割图像。通过这种方法，我们在马萨诸岛建筑和路径数据集上达到了 benchmark 分数，具体分别为0.8365和0.9153。
Digital Twins for Patient Care via Knowledge Graphs and Closed-Form Continuous-Time Liquid Neural Networks
results: 本研究的结果表明，使用这种新的框架可以实现实时的患者健康状况概述、个性化医疗和早期诊断、模拟治疗结果和优化手术规划，为数字双技术在医疗领域的应用提供了新的可能性。Abstract
Digital twin technology has is anticipated to transform healthcare, enabling personalized medicines and support, earlier diagnoses, simulated treatment outcomes, and optimized surgical plans. Digital twins are readily gaining traction in industries like manufacturing, supply chain logistics, and civil infrastructure. Not in patient care, however. The challenge of modeling complex diseases with multimodal patient data and the computational complexities of analyzing it have stifled digital twin adoption in the biomedical vertical. Yet, these major obstacles can potentially be handled by approaching these models in a different way. This paper proposes a novel framework for addressing the barriers to clinical twin modeling created by computational costs and modeling complexities. We propose structuring patient health data as a knowledge graph and using closed-form continuous-time liquid neural networks, for real-time analytics. By synthesizing multimodal patient data and leveraging the flexibility and efficiency of closed form continuous time networks and knowledge graph ontologies, our approach enables real time insights, personalized medicine, early diagnosis and intervention, and optimal surgical planning. This novel approach provides a comprehensive and adaptable view of patient health along with real-time analytics, paving the way for digital twin simulations and other anticipated benefits in healthcare.
摘要
《数字双胞技术在医疗领域的应用》数字双胞技术已被预测将重点改变医疗领域，使得个性化药物和支持、早期诊断、模拟治疗结果和优化手术规划等became possible。然而，在病人护理方面，数字双胞技术的普及受到了许多挑战，主要是模拟复杂的疾病和多模态病人数据的计算复杂性。然而，这些主要障碍可以通过一种新的方法来解决。本文提出了一种新的框架，用于解决在计算成本和模型复杂性等方面阻碍临床双胞模型的障碍。我们提议将病人健康数据结构化为知识图，并使用闭合形时间连续神经网络，以实现实时分析。通过对多模态病人数据进行synthesize，并利用closed-form continuous-time liquid neural networks和知识图 ontologies的灵活性和高效性，我们的方法可以实现实时的医疗探索和个性化医疗。本文的新方法可以为医疗领域提供一个普适和可变的病人健康视图，同时实现实时的分析，为数字双胞 simulations和其他预期的健康医疗带来了新的机遇。
Fault Monitoring in Passive Optical Networks using Machine Learning Techniques
methods: 利用机器学习（ML）技术进行PON系统故障监测，并通过实验Optical Time Domain Reflectometry（OTDR）数据验证其效果。
results: 通过ML方法实现PON系统故障监测，可以减少服务中断时的财务损失风险，并提高系统的可靠性和稳定性。Abstract
Passive optical network (PON) systems are vulnerable to a variety of failures, including fiber cuts and optical network unit (ONU) transmitter/receiver failures. Any service interruption caused by a fiber cut can result in huge financial losses for service providers or operators. Identifying the faulty ONU becomes difficult in the case of nearly equidistant branch terminations because the reflections from the branches overlap, making it difficult to distinguish the faulty branch given the global backscattering signal. With increasing network size, the complexity of fault monitoring in PON systems increases, resulting in less reliable monitoring. To address these challenges, we propose in this paper various machine learning (ML) approaches for fault monitoring in PON systems, and we validate them using experimental optical time domain reflectometry (OTDR) data.
摘要
激光网络（PON）系统受到多种故障的威胁，包括纤维断和Optical Network Unit（ONU）发送器/接收器故障。任何纤维断导致的服务中断可能会对服务提供商或运营商造成巨大的经济损失。在分支结束处 nearly equidistant的情况下，缺陷的ONU diffficult to identify，因为分支 reflection overlap，使得不可以通过全球反射信号来 отличи出缺陷分支。随着网络规模的增加，PON系统中的缺陷监测复杂度增加，导致监测变得更加不可靠。为解决这些挑战，本文提出了基于机器学习（ML）的多种缺陷监测方法，并通过实验optical time domain reflectometry（OTDR）数据 validate them。
Rosko: Row Skipping Outer Products for Sparse Matrix Multiplication Kernels
results: Rosko kernels可以与其他outer product scheduling方法结合使用，并且可以将其他方法的计算 skipped by using Rosko的packing format。Rosko kernels在实际硬件上比 EXISTS auto-tuning和搜索基于解决方案和商业供应链的 vendor-optimized 库来的性能更好，并且可以在不同的神经网络负载上实现更高的执行速度，达到6.5倍的时间优化。Abstract
We propose Rosko -- row skipping outer products -- for deriving sparse matrix multiplication (SpMM) kernels in reducing computation and memory access requirements of deep neural networks (DNNs). Rosko allows skipping of entire row computations during program execution with low sparsity-management overheads. We analytically derive sparse CPU kernels that adapt to given hardware characteristics to effectively utilize processor cores and minimize data movement without the need for auto-tuning or search space exploration. Rosko can be integrated with other outer product scheduling methods, allowing them to leverage row skipping by using Rosko's packing format to skip unnecessary computation. Rosko kernels outperform existing auto-tuning and search-based solutions as well as state-of-the-art vendor-optimized libraries on real hardware across a variety of neural network workloads. For matrices with sparsities ranging from 65% to 99.8% typically found in machine learning, Rosko kernels achieve up to a 6.5x runtime reduction on Intel and ARM CPUs.
摘要
我们提出了Rosko，它是跳过外积栈的方法，用于获得深度神经网络（DNN）中的简约矩阵乘法（SpMM）内核。Rosko可以在程式执行时跳过整个行的计算，具有低简约管理成本。我们分析性地 derivatesparse CPU内核，可以根据硬件特性来有效利用处理器核心和减少资料移动，无需进行自动调整或搜索空间探索。Rosko可以与其他外积栈调度方法结合，让它们利用Rosko的封包格式跳过无需计算。Rosko内核比现有的自动调整和搜索基于解决方案以及商业化优化库在真实硬件上表现更好，对于具有65%到99.8%的简约率，通常见于机器学习 tasks 中，Rosko内核在英特尔和ARM CPU上可以获得至多6.5倍的执行时间优化。
paper_authors: April Chen, Ryan A. Rossi, Namyong Park, Puja Trivedi, Yu Wang, Tong Yu, Sungchul Kim, Franck Dernoncourt, Nesreen K. Ahmed for: This paper focuses on improving the fairness of Graph Neural Networks (GNNs) by examining and categorizing fairness techniques for GNNs.methods: The paper discusses previous work on fair GNN models and techniques, including those that focus on improving fairness during preprocessing, training, or post-processing. The paper also introduces an intuitive taxonomy for fairness evaluation metrics.results: The paper highlights the advantages and intuition of using fairness techniques in GNNs, and summarizes graph datasets that are useful for benchmarking the fairness of GNN models. The paper also identifies key open problems and challenges that remain to be addressed.Abstract
Graph Neural Networks (GNNs) have become increasingly important due to their representational power and state-of-the-art predictive performance on many fundamental learning tasks. Despite this success, GNNs suffer from fairness issues that arise as a result of the underlying graph data and the fundamental aggregation mechanism that lies at the heart of the large class of GNN models. In this article, we examine and categorize fairness techniques for improving the fairness of GNNs. Previous work on fair GNN models and techniques are discussed in terms of whether they focus on improving fairness during a preprocessing step, during training, or in a post-processing phase. Furthermore, we discuss how such techniques can be used together whenever appropriate, and highlight the advantages and intuition as well. We also introduce an intuitive taxonomy for fairness evaluation metrics including graph-level fairness, neighborhood-level fairness, embedding-level fairness, and prediction-level fairness metrics. In addition, graph datasets that are useful for benchmarking the fairness of GNN models are summarized succinctly. Finally, we highlight key open problems and challenges that remain to be addressed.
摘要
graph neural networks (GNNs) 在最近几年内变得越来越重要，这是因为它们在许多基本学习任务上具有代表力和状态速度。然而，GNNs 受到公平问题的影响，这些问题 arise 由下面的图数据和基本聚合机制。在这篇文章中，我们评估和分类 fairness 技术，以提高 GNNs 的公平性。先前的 fair GNN 模型和技术被分为三类：在预处理阶段、在训练阶段和在后处理阶段。此外，我们还讨论了这些技术可以在合适的时候一起使用，并高亮了其优势和直觉。此外，我们还提出了一个直观的公平评价度量分类，包括图级公平、邻居级公平、嵌入级公平和预测级公平度量。此外，我们还简要介绍了一些用于评估 GNN 模型公平性的图数据。最后，我们高亮了一些未解决的问题和挑战。Note: The translation is in Simplified Chinese, which is the standard writing system used in mainland China. If you need the translation in Traditional Chinese, please let me know.
results: 对比较几种现有算法，提出的两种算法具有高效计算和可靠的enario-based模型特点，并在股票投资中得到了广泛的应用。Abstract
We seek to extract a small number of representative scenarios from large and high-dimensional panel data that are consistent with sample moments. Among two novel algorithms, the first identifies scenarios that have not been observed before, and comes with a scenario-based representation of covariance matrices. The second proposal picks important data points from states of the world that have already realized, and are consistent with higher-order sample moment information. Both algorithms are efficient to compute, and lend themselves to consistent scenario-based modeling and high-dimensional numerical integration. Extensive numerical benchmarking studies and an application in portfolio optimization favor the proposed algorithms.
摘要
我们寻求从大量高维批处理数据中提取一小量表示性的情景，这些情景与样本幂应于一致。我们提出了两种新算法，第一种可以找到没有被观察到的情景，同时提供了情景基于协方差矩阵的表示方式。第二种算法选择已经实现的世界状态中重要的数据点，并与高阶样本幂信息一致。这两种算法都具有计算效率，适用于一致的enario-based模型和高维数字 интегра。我们在大量计算和股票投资应用中进行了广泛的数值对比研究，而这两种算法都得到了 preference。
Training Physics-Informed Neural Networks via Multi-Task Optimization for Traffic Density Prediction
paper_authors: Bo Wang, A. K. Qin, Sajjad Shafiei, Hussein Dia, Adriana-Simona Mihaita, Hanna Grzybowska
for: 用于预测交通流量
methods: 使用多任务优化（MTO）框架，创建多个辅助任务并与主任务一起解决
results: 实验结果表明，我们提议的训练框架可以在比较限制的数据量下提高PINN的性能Abstract
Physics-informed neural networks (PINNs) are a newly emerging research frontier in machine learning, which incorporate certain physical laws that govern a given data set, e.g., those described by partial differential equations (PDEs), into the training of the neural network (NN) based on such a data set. In PINNs, the NN acts as the solution approximator for the PDE while the PDE acts as the prior knowledge to guide the NN training, leading to the desired generalization performance of the NN when facing the limited availability of training data. However, training PINNs is a non-trivial task largely due to the complexity of the loss composed of both NN and physical law parts. In this work, we propose a new PINN training framework based on the multi-task optimization (MTO) paradigm. Under this framework, multiple auxiliary tasks are created and solved together with the given (main) task, where the useful knowledge from solving one task is transferred in an adaptive mode to assist in solving some other tasks, aiming to uplift the performance of solving the main task. We implement the proposed framework and apply it to train the PINN for addressing the traffic density prediction problem. Experimental results demonstrate that our proposed training framework leads to significant performance improvement in comparison to the traditional way of training the PINN.
摘要
Physics-informed neural networks (PINNs) 是一个新兴的研究领域，它将物理法则 incorporated 到 neural network (NN) 的训练中，使得 NN 能够更好地预测数据集中的特征。在 PINNs 中，NN acted as 数据集中的解决方案，而 PDE acted as 导航 NN 训练的知识。这使得 NN 在面临有限的训练数据时能够达到更好的总体性能。然而，训练 PINNs 是一个非rivial任务，主要因为损失函数的复杂性，它包括 NN 和物理法则部分。在这种情况下，我们提出了一个基于多任务优化 (MTO) 的新的训练框架。在这个框架中，我们创建了多个auxiliary任务，并与主任务一起解决，使得解决一个任务的有用知识可以在适应模式下传递给另一个任务，以提高主任务的解决性能。我们实现了提议的框架，并应用它来训练 PINN 来解决交通密度预测问题。实验结果表明，我们的训练框架可以在传统训练 PINN 的基础上获得显著性能提升。
Incorporating Deep Q – Network with Multiclass Classification Algorithms
results: 本研究发现，通过使用DQN，可以提高多类分类精度和稳定性。这些结果可以用于各种领域，包括图像识别、自然语言处理和生物信息学。特别是在金融风险管理和财务预测方面，可以用DQN来预测企业面临financial distress的可能性。Abstract
In this study, we explore how Deep Q-Network (DQN) might improve the functionality of multiclass classification algorithms. We will use a benchmark dataset from Kaggle to create a framework incorporating DQN with existing supervised multiclass classification algorithms. The findings of this study will bring insight into how deep reinforcement learning strategies may be used to increase multiclass classification accuracy. They have been used in a number of fields, including image recognition, natural language processing, and bioinformatics. This study is focused on the prediction of financial distress in companies in addition to the wider application of Deep Q-Network in multiclass classification. Identifying businesses that are likely to experience financial distress is a crucial task in the fields of finance and risk management. Whenever a business experiences serious challenges keeping its operations going and meeting its financial responsibilities, it is said to be in financial distress. It commonly happens when a company has a sharp and sustained recession in profitability, cash flow issues, or an unsustainable level of debt.
摘要
在这个研究中，我们探讨了深度Q网络（DQN）如何改善多类分类算法的功能。我们将使用Kaggle的标准数据集创建一个框架，该框架将包含DQN与现有的多类分类算法。我们的发现将提供深入理解deep reinforcement learning策略如何提高多类分类精度。这些策略在图像识别、自然语言处理和生物信息处理等领域都有广泛的应用。本研究的特点是用DQN预测公司面临财务危机的可能性，而不是仅仅是将其应用于多类分类。公司经历严重的运营困难和履行财务责任时，被称为财务危机。这通常发生在公司收入下降、财务流动性困难或债务水平不可持续时。
ScriptWorld: Text Based Environment For Learning Procedural Knowledge
results: 实验表明，基于预训练语言模型的语言特征可以帮助RL算法解决日常生活中的文本基础环境问题。Abstract
Text-based games provide a framework for developing natural language understanding and commonsense knowledge about the world in reinforcement learning based agents. Existing text-based environments often rely on fictional situations and characters to create a gaming framework and are far from real-world scenarios. In this paper, we introduce ScriptWorld: a text-based environment for teaching agents about real-world daily chores and hence imparting commonsense knowledge. To the best of our knowledge, it is the first interactive text-based gaming framework that consists of daily real-world human activities designed using scripts dataset. We provide gaming environments for 10 daily activities and perform a detailed analysis of the proposed environment. We develop RL-based baseline models/agents to play the games in Scriptworld. To understand the role of language models in such environments, we leverage features obtained from pre-trained language models in the RL agents. Our experiments show that prior knowledge obtained from a pre-trained language model helps to solve real-world text-based gaming environments. We release the environment via Github: https://github.com/Exploration-Lab/ScriptWorld
摘要
Feature selection simultaneously preserving both class and cluster structures
results: 实验结果表明，提议的特征/带选择方法可以选择一 subset of 特征，这些特征是良好的分类和归类。Abstract
When a data set has significant differences in its class and cluster structure, selecting features aiming only at the discrimination of classes would lead to poor clustering performance, and similarly, feature selection aiming only at preserving cluster structures would lead to poor classification performance. To the best of our knowledge, a feature selection method that simultaneously considers class discrimination and cluster structure preservation is not available in the literature. In this paper, we have tried to bridge this gap by proposing a neural network-based feature selection method that focuses both on class discrimination and structure preservation in an integrated manner. In addition to assessing typical classification problems, we have investigated its effectiveness on band selection in hyperspectral images. Based on the results of the experiments, we may claim that the proposed feature/band selection can select a subset of features that is good for both classification and clustering.
摘要
当数据集具有显著的类和差异结构时，仅仅选择用于分类的特征会导致群集性能差，而仅仅保持群集结构的特征选择也会导致分类性能差。据我们所知，Literature中没有一种同时考虑类分化和群集结构保持的特征选择方法。在这篇论文中，我们尝试了bridging这个差距，提出了基于神经网络的特征选择方法，该方法同时考虑类分化和群集结构保持。除了评估典型的分类问题之外，我们还对带选择在多spectral图像进行了研究。根据实验结果，我们可以确认，我们提议的特征/带选择方法可以选择一个良好的分类和群集结构的子集。
Active Learning in Physics: From 101, to Progress, and Perspective
results: 这篇论文提出了将活动学习与量子机器学习（QL）融合的想法，以实现这两个领域之间的共融。Abstract
Active Learning (AL) is a family of machine learning (ML) algorithms that predates the current era of artificial intelligence. Unlike traditional approaches that require labeled samples for training, AL iteratively selects unlabeled samples to be annotated by an expert. This protocol aims to prioritize the most informative samples, leading to improved model performance compared to training with all labeled samples. In recent years, AL has gained increasing attention, particularly in the field of physics. This paper presents a comprehensive and accessible introduction to the theory of AL reviewing the latest advancements across various domains. Additionally, we explore the potential integration of AL with quantum ML, envisioning a synergistic fusion of these two fields rather than viewing AL as a mere extension of classical ML into the quantum realm.
摘要
aktiv learning (AL) 是一家机器学习（ML）算法家族，比现代人工智能更早出现。不同于传统的方法，AL 在训练过程中不需要标注样本，而是逐渐选择无标注样本，由专家进行标注。这个协议的目的是优化模型性能，与全部标注样本训练相比。在最近几年，AL 在物理领域获得了越来越多的注意，特别是在物理领域。这篇论文提供了 AL 的完整和可访问的理论介绍，同时还探讨了 AL 与量子 ML 的可能的集成，而不是视 AL 为класси ML 在量子世界的扩展。
Incomplete Utterance Rewriting as Sequential Greedy Tagging
results: 我们的模型在多个公共数据集上实现了最佳的 nine restoration scores，而其他 metric scores 与之前的状态OF-the-art模型相比可观。此外，由于我们的模型简单，在推理速度方面也超过了大多数之前的模型。Abstract
The task of incomplete utterance rewriting has recently gotten much attention. Previous models struggled to extract information from the dialogue context, as evidenced by the low restoration scores. To address this issue, we propose a novel sequence tagging-based model, which is more adept at extracting information from context. Meanwhile, we introduce speaker-aware embedding to model speaker variation. Experiments on multiple public datasets show that our model achieves optimal results on all nine restoration scores while having other metric scores comparable to previous state-of-the-art models. Furthermore, benefitting from the model's simplicity, our approach outperforms most previous models on inference speed.
摘要
“ incomplete utterance rewriting ”在最近已经获得了很多注意。以前的模型很难从对话 контекст中提取信息，这可以见到低恢复得分。为解决这个问题，我们提议一个新的序列标签基于模型，这个模型更能够从 контекст中提取信息。同时，我们引入了Speaker-aware embedding来模型说话者的变化。多个公共数据集上的实验表明，我们的模型在所有九个恢复得分上取得了最佳结果，而其他指标得分与过往最佳模型相近。此外，由于我们的方法简单，我们的方法在推理速度方面超越了大多数前一代模型。
Improving Prototypical Part Networks with Reward Reweighing, Reselection, and Retraining
for: 该paper aimed to improve the interpretability of deep learning models for image classification by using human feedback to fine-tune the prototypes.
methods: 该paper proposed a novel method called R3-ProtoPNet, which combines reward-based reweighting, reselection, and retraining to align the model’s features with the updated prototypes.
results: 该paper found that R3-ProtoPNet improves the overall consistency and meaningfulness of the prototypes, but lower the test predictive accuracy when used independently. However, when multiple R3-ProtoPNets are incorporated into an ensemble, the test predictive performance is increased while maintaining interpretability.Abstract
In recent years, work has gone into developing deep interpretable methods for image classification that clearly attributes a model's output to specific features of the data. One such of these methods is the prototypical part network (ProtoPNet), which attempts to classify images based on meaningful parts of the input. While this method results in interpretable classifications, this method often learns to classify from spurious or inconsistent parts of the image. Hoping to remedy this, we take inspiration from the recent developments in Reinforcement Learning with Human Feedback (RLHF) to fine-tune these prototypes. By collecting human annotations of prototypes quality via a 1-5 scale on the CUB-200-2011 dataset, we construct a reward model that learns to identify non-spurious prototypes. In place of a full RL update, we propose the reweighted, reselected, and retrained prototypical part network (R3-ProtoPNet), which adds an additional three steps to the ProtoPNet training loop. The first two steps are reward-based reweighting and reselection, which align prototypes with human feedback. The final step is retraining to realign the model's features with the updated prototypes. We find that R3-ProtoPNet improves the overall consistency and meaningfulness of the prototypes, but lower the test predictive accuracy when used independently. When multiple R3-ProtoPNets are incorporated into an ensemble, we find an increase in test predictive performance while maintaining interpretability.
摘要
recent years, work has gone into developing deep interpretable methods for image classification that clearly attributes a model's output to specific features of the data. One such of these methods is the prototypical part network (ProtoPNet), which attempts to classify images based on meaningful parts of the input. While this method results in interpretable classifications, this method often learns to classify from spurious or inconsistent parts of the image. Hoping to remedy this, we take inspiration from the recent developments in Reinforcement Learning with Human Feedback (RLHF) to fine-tune these prototypes. By collecting human annotations of prototypes quality via a 1-5 scale on the CUB-200-2011 dataset, we construct a reward model that learns to identify non-spurious prototypes. In place of a full RL update, we propose the reweighted, reselected, and retrained prototypical part network (R3-ProtoPNet), which adds an additional three steps to the ProtoPNet training loop. The first two steps are reward-based reweighting and reselection, which align prototypes with human feedback. The final step is retraining to realign the model's features with the updated prototypes. We find that R3-ProtoPNet improves the overall consistency and meaningfulness of the prototypes, but lower the test predictive accuracy when used independently. When multiple R3-ProtoPNets are incorporated into an ensemble, we find an increase in test predictive performance while maintaining interpretability.Here's the translation in Traditional Chinese:在近年来，有很多工作在开发深度可解释的图像分类方法，以将模型的输出明确地对对应的资料特征进行推导。一个如此方法是 прототипіаль部分网络（ProtoPNet），它尝试根据输入图像的意义部分进行分类。although this method results in interpretable classifications, this method often learns to classify from spurious or inconsistent parts of the image. hoping to remedy this, we take inspiration from the recent developments in Reinforcement Learning with Human Feedback (RLHF) to fine-tune these prototypes. by collecting human annotations of prototypes quality via a 1-5 scale on the CUB-200-2011 dataset, we construct a reward model that learns to identify non-spurious prototypes. in place of a full RL update, we propose the reweighted, reselected, and retrained prototypical part network (R3-ProtoPNet), which adds an additional three steps to the ProtoPNet training loop. the first two steps are reward-based reweighting and reselection, which align prototypes with human feedback. the final step is retraining to realign the model's features with the updated prototypes. we find that R3-ProtoPNet improves the overall consistency and meaningfulness of the prototypes, but lower the test predictive accuracy when used independently. when multiple R3-ProtoPNets are incorporated into an ensemble, we find an increase in test predictive performance while maintaining interpretability.
On Regularization and Inference with Label Constraints
results: 论文表明，正则化可以减少泛化差距，但是它会偏好小违反，导致模型偏离优质点。而受约束推理则可以降低人口风险，从而使得违反变成了优势。因此，论文建议在使用这两种方法时，可以共同使用它们，并在某些条件下使用约束推理来补偿正则化引入的偏见。Abstract
Prior knowledge and symbolic rules in machine learning are often expressed in the form of label constraints, especially in structured prediction problems. In this work, we compare two common strategies for encoding label constraints in a machine learning pipeline, regularization with constraints and constrained inference, by quantifying their impact on model performance. For regularization, we show that it narrows the generalization gap by precluding models that are inconsistent with the constraints. However, its preference for small violations introduces a bias toward a suboptimal model. For constrained inference, we show that it reduces the population risk by correcting a model's violation, and hence turns the violation into an advantage. Given these differences, we further explore the use of two approaches together and propose conditions for constrained inference to compensate for the bias introduced by regularization, aiming to improve both the model complexity and optimal risk.
摘要
Prior knowledge和符号规则在机器学习中经常表达为标签约束，特别是在结构预测问题中。在这项工作中，我们比较了两种常见的标签约束编码策略在机器学习管道中的影响，即规范减少和受约束的推理。对于规范，我们表明了它可以防止模型与约束不一致的情况，从而缩小泛化差。但是，它会偏好小规模的违反，导致模型偏好一个不佳的模型。对于受约束推理，我们表明了它可以降低人口风险，通过约束违反的修正，使违反变成一个优势。基于这些差异，我们进一步探索了两种方法的同时使用，并提出了限制推理可以资COMPENSATE FOR THE BIAS INTRODUCED BY REGULARIZATION，以提高模型复杂度和优化风险。
Noisy Tensor Ring approximation for computing gradients of Variational Quantum Eigensolver for Combinatorial Optimization
results: 比较分类计算和量子计算的复杂度，显示这种方法可以减少分类计算的复杂度，使其可以更快地评估量子算法的梯度。Abstract
Variational Quantum algorithms, especially Quantum Approximate Optimization and Variational Quantum Eigensolver (VQE) have established their potential to provide computational advantage in the realm of combinatorial optimization. However, these algorithms suffer from classically intractable gradients limiting the scalability. This work addresses the scalability challenge for VQE by proposing a classical gradient computation method which utilizes the parameter shift rule but computes the expected values from the circuits using a tensor ring approximation. The parametrized gates from the circuit transform the tensor ring by contracting the matrix along the free edges of the tensor ring. While the single qubit gates do not alter the ring structure, the state transformations from the two qubit rotations are evaluated by truncating the singular values thereby preserving the structure of the tensor ring and reducing the computational complexity. This variation of the Matrix product state approximation grows linearly in number of qubits and the number of two qubit gates as opposed to the exponential growth in the classical simulations, allowing for a faster evaluation of the gradients on classical simulators.
摘要
“精简量子算法，尤其是量子近似优化和量子对角器（VQE），已经证明了它们在排序问题中的计算优势。然而，这些算法受到古典无法计算的梯度所限制，这使得扩展性受到挑战。本工作解决了VQE的扩展性问题，提出一种古典梯度计算方法，利用参数移动规则，并从图静态环节中计算出预期值。图静态环节中的参数门由图静态环节中的矩阵折缩，单位门不改变环节结构，但两个量子矩阵的状态转换则被舒缓，以保持环节结构并降低计算复杂性。这种矩阵产品state的扩展增长Linearly在量子矩阵中，相比之下，古典 simulations中的扩展增长 exponential，使得在古典 simulators 上较快地评估梯度。”
Large Language Models for Supply Chain Optimization
for: The paper is written for supply chain operators and managers who need to interpret and explain the outcomes of optimization algorithms to stakeholders.
methods: The paper proposes a framework called OptiGuide that leverages Large Language Models (LLMs) to provide insights into the underlying optimization outcomes. The framework accepts queries in plain text and outputs explanations of the optimization results without requiring the transfer of proprietary data to the LLM.
results: The paper demonstrates the effectiveness of OptiGuide on a real server placement scenario within Microsoft’s cloud supply chain. The results show that OptiGuide can provide accurate explanations of the optimization outcomes, and the proposed evaluation benchmark can be used to evaluate the accuracy of the LLM output in other scenarios.Abstract
Supply chain operations traditionally involve a variety of complex decision making problems. Over the last few decades, supply chains greatly benefited from advances in computation, which allowed the transition from manual processing to automation and cost-effective optimization. Nonetheless, business operators still need to spend substantial efforts in explaining and interpreting the optimization outcomes to stakeholders. Motivated by the recent advances in Large Language Models (LLMs), we study how this disruptive technology can help bridge the gap between supply chain automation and human comprehension and trust thereof. We design OptiGuide -- a framework that accepts as input queries in plain text, and outputs insights about the underlying optimization outcomes. Our framework does not forgo the state-of-the-art combinatorial optimization technology, but rather leverages it to quantitatively answer what-if scenarios (e.g., how would the cost change if we used supplier B instead of supplier A for a given demand?). Importantly, our design does not require sending proprietary data over to LLMs, which can be a privacy concern in some circumstances. We demonstrate the effectiveness of our framework on a real server placement scenario within Microsoft's cloud supply chain. Along the way, we develop a general evaluation benchmark, which can be used to evaluate the accuracy of the LLM output in other scenarios.
摘要
供应链运营传统上涉及到许多复杂的决策问题。过去几十年，供应链受计算技术的进步所助，从人工处理过渡到自动化和成本效果优化。然而，业务运营者仍需投入很大的努力来解释和理解优化结果，以获得投资者和客户的信任。鼓励于最近的大语言模型（LLM）技术的进步，我们研究如何使用这种破坏技术来bridging供应链自动化和人类理解之间的差距。我们设计了OptiGuide框架，它可以接受普通文本查询，并输出供应链优化结果的概念。我们的框架不会抛弃现有的组合优化技术，而是利用它们来回答什么时候的问题（例如，如果使用供应商B而不是供应商A来满足某个需求时，成本会如何变化？）。重要的是，我们的设计不需要将专有数据传递到LLM中，这可能会在某些情况下成为隐私问题。我们在微软云供应链中进行了实际的服务分配enario，并在过程中开发了一个通用评估标准，可以用于评估其他场景中LLM输出的准确性。
Domain Adaptation using Silver Standard Labels for Ki-67 Scoring in Digital Pathology: A Step Closer to Widescale Deployment
paper_authors: Amanda Dy, Ngoc-Nhu Jennifer Nguyen, Seyed Hossein Mirjahanmardi, Melanie Dawe, Anthony Fyles, Wei Shi, Fei-Fei Liu, Dimitrios Androutsos, Susan Done, April Khademi for: 这个研究旨在提高 Ki-67 PI 分配的 объектив性和效率，使用深度学习系统。methods: 该研究提出了一个领域适应管道，使用无监督框架生成目标领域的银标签，以增强源频率银标签数据的学习效果。results: 比较 SS Only、GS Only、Mixed、GS+SS 和我们的提议方法 SS+GS 的五种训练方案，SS+GS 方法在目标数据上显示出最高的 PI 准确率（95.9%）和更一致的结果，与 GS Only 模型在目标数据上的表现有 statistically significant difference（p < 0.05）。Abstract
Deep learning systems have been proposed to improve the objectivity and efficiency of Ki- 67 PI scoring. The challenge is that while very accurate, deep learning techniques suffer from reduced performance when applied to out-of-domain data. This is a critical challenge for clinical translation, as models are typically trained using data available to the vendor, which is not from the target domain. To address this challenge, this study proposes a domain adaptation pipeline that employs an unsupervised framework to generate silver standard (pseudo) labels in the target domain, which is used to augment the gold standard (GS) source domain data. Five training regimes were tested on two validated Ki-67 scoring architectures (UV-Net and piNET), (1) SS Only: trained on target silver standard (SS) labels, (2) GS Only: trained on source GS labels, (3) Mixed: trained on target SS and source GS labels, (4) GS+SS: trained on source GS labels and fine-tuned on target SS labels, and our proposed method (5) SS+GS: trained on source SS labels and fine-tuned on source GS labels. The SS+GS method yielded significantly (p < 0.05) higher PI accuracy (95.9%) and more consistent results compared to the GS Only model on target data. Analysis of t-SNE plots showed features learned by the SS+GS models are more aligned for source and target data, resulting in improved generalization. The proposed pipeline provides an efficient method for learning the target distribution without manual annotations, which are time-consuming and costly to generate for medical images. This framework can be applied to any target site as a per-laboratory calibration method, for widescale deployment.
摘要
深度学习系统已被提议以提高基因67PI分配的 объектив性和效率。然而，深度学习技术在域外数据上表现不佳是一个重要挑战，这是因为模型通常在生产商提供的数据上进行训练，而不是target域的数据。为解决这个挑战，本研究提出了一个适应域pipeline，该pipeline使用了无监督框架生成target域的银标(pseudo)标签，并将其用于增强来自源域的金标(GS)数据。本研究测试了五种训练方案，包括(1) SS Only：基于target银标(SS)标签进行训练，(2) GS Only：基于源GS标签进行训练，(3) Mixed：基于target SS和源GS标签进行训练，(4) GS+SS：基于源GS标签进行训练，并在target SS标签上进行微调，以及我们的提议方法(5) SS+GS：基于源SS标签进行训练，并在源GS标签上进行微调。SS+GS方法在target数据上得到了 statistically significant (p < 0.05) 的PI准确率（95.9%），并且与源数据的结果更一致。分析t-SNE图表显示，SS+GS模型学习的特征更加适应于源和目标数据，导致了改善的总体性。本ipeline提供了一种效率的方法来学习目标分布，不需要手动生成昂贵和时间consuming的医学图像标签。这种框架可以在任何目标站点上应用，作为室内准确方法进行大规模部署。
When Do Transformers Shine in RL? Decoupling Memory from Credit Assignment
methods: 该研究使用了Formal definitions of memory length和credit assignment length来测试Transformer-based RL方法的表现。
results: 研究发现，Transformers可以增强RL算法的记忆能力，可以扩展到需要记忆步骤1500个的任务。但是，Transformers不会改善长期归因。Abstract
Reinforcement learning (RL) algorithms face two distinct challenges: learning effective representations of past and present observations, and determining how actions influence future returns. Both challenges involve modeling long-term dependencies. The transformer architecture has been very successful to solve problems that involve long-term dependencies, including in the RL domain. However, the underlying reason for the strong performance of Transformer-based RL methods remains unclear: is it because they learn effective memory, or because they perform effective credit assignment? After introducing formal definitions of memory length and credit assignment length, we design simple configurable tasks to measure these distinct quantities. Our empirical results reveal that Transformers can enhance the memory capacity of RL algorithms, scaling up to tasks that require memorizing observations $1500$ steps ago. However, Transformers do not improve long-term credit assignment. In summary, our results provide an explanation for the success of Transformers in RL, while also highlighting an important area for future research and benchmark design.
摘要
reinforcement learning (RL) 算法面临两个不同的挑战：学习过去和当前观察的有效表示，以及确定行动对未来返回的影响。两个挑战都涉及到模型长期关系。 transformer 架构在RL领域中具有非常出色的表现，但是下面的原因仍然不清楚：是因为它们学习有效的记忆，或者是因为它们实现有效的准确分配？我们给出了正式的定义 memory length 和 credit assignment length，然后设计了简单可配置的任务来测量这两个特点。我们的实验结果表明，Transformers 可以增强 RL 算法的记忆容量，可以扩展到需要记忆 observation 1500 步的任务。但是，Transformers 不会改善长期准确分配。简单来说，我们的结果可以解释 transformer 在 RL 中的成功，同时也提出了未来研究和标准化的测试设计。
Memory-Immersed Collaborative Digitization for Area-Efficient Compute-in-Memory Deep Learning
results: 这个研究使用了65奈米CMOS试验板，与40奈米节点5位SAR ADC和40奈米节点5位Flash ADC进行比较，结果显示，这个设计需要相对于40奈米节点SAR ADC的面积和能源减少为$\sim$25$\times$和$\sim$1.4$\times$，相对于40奈米节点Flash ADC的面积和能源减少为$\sim$51$\times$和$\sim$13$\times$。Abstract
This work discusses memory-immersed collaborative digitization among compute-in-memory (CiM) arrays to minimize the area overheads of a conventional analog-to-digital converter (ADC) for deep learning inference. Thereby, using the proposed scheme, significantly more CiM arrays can be accommodated within limited footprint designs to improve parallelism and minimize external memory accesses. Under the digitization scheme, CiM arrays exploit their parasitic bit lines to form a within-memory capacitive digital-to-analog converter (DAC) that facilitates area-efficient successive approximation (SA) digitization. CiM arrays collaborate where a proximal array digitizes the analog-domain product-sums when an array computes the scalar product of input and weights. We discuss various networking configurations among CiM arrays where Flash, SA, and their hybrid digitization steps can be efficiently implemented using the proposed memory-immersed scheme. The results are demonstrated using a 65 nm CMOS test chip. Compared to a 40 nm-node 5-bit SAR ADC, our 65 nm design requires $\sim$25$\times$ less area and $\sim$1.4$\times$ less energy by leveraging in-memory computing structures. Compared to a 40 nm-node 5-bit Flash ADC, our design requires $\sim$51$\times$ less area and $\sim$13$\times$ less energy.
摘要
In the proposed digitization scheme, CiM arrays utilize their parasitic bit lines to form a within-memory capacitive digital-to-analog converter (DAC) that enables area-efficient successive approximation (SA) digitization. CiM arrays collaborate by digitizing the analog-domain product-sums when one array computes the scalar product of input and weights.The proposed memory-immersed scheme can efficiently implement various networking configurations among CiM arrays, including Flash, SA, and their hybrid digitization steps. The results are demonstrated using a 65 nm CMOS test chip. Compared to a 40 nm-node 5-bit SAR ADC, our 65 nm design requires approximately 25 times less area and 1.4 times less energy. Compared to a 40 nm-node 5-bit Flash ADC, our design requires approximately 51 times less area and 13 times less energy.
A Natural Language Processing Approach to Malware Classification
results: 研究发现，使用这种NLP基于的方法可以在一个复杂的恶意软件集合上达到最佳效果，HMM-Random Forrest模型在这个集合上得到了最佳结果。Abstract
Many different machine learning and deep learning techniques have been successfully employed for malware detection and classification. Examples of popular learning techniques in the malware domain include Hidden Markov Models (HMM), Random Forests (RF), Convolutional Neural Networks (CNN), Support Vector Machines (SVM), and Recurrent Neural Networks (RNN) such as Long Short-Term Memory (LSTM) networks. In this research, we consider a hybrid architecture, where HMMs are trained on opcode sequences, and the resulting hidden states of these trained HMMs are used as feature vectors in various classifiers. In this context, extracting the HMM hidden state sequences can be viewed as a form of feature engineering that is somewhat analogous to techniques that are commonly employed in Natural Language Processing (NLP). We find that this NLP-based approach outperforms other popular techniques on a challenging malware dataset, with an HMM-Random Forrest model yielding the best results.
摘要
很多不同的机器学习和深度学习技术已经成功地应用于恶意软件检测和分类。例如，常见的学习技术在恶意软件领域包括隐藏markov模型（HMM）、Random Forests（RF）、卷积神经网络（CNN）、支持向量机器（SVM）和回归神经网络（RNN），如长期短时间记忆网络（LSTM）。在这些研究中，我们考虑了一种混合体系，其中HMM被训练于机器码序列，并将这些训练得到的隐藏状态用作不同的分类器的特征向量。在这种情况下，提取HMM隐藏状态序列可以被视为一种特征工程技术，与自然语言处理（NLP）中常见的技术有一定的相似性。我们发现，这种NLP基于的方法在一个复杂的恶意软件数据集上表现出色，HMM-Random Forrest模型实现了最佳结果。
results: 这个研究获得了0.78的分类精度（即用户识别率），但是使用Random Forest分类器并不对相似的特征进行轻微修改后，获得了0.93的分类精度。Abstract
In previous research, keystroke dynamics has shown promise for user authentication, based on both fixed-text and free-text data. In this research, we consider the more challenging multiclass user identification problem, based on free-text data. We experiment with a complex image-like feature that has previously been used to achieve state-of-the-art authentication results over free-text data. Using this image-like feature and multiclass Convolutional Neural Networks, we are able to obtain a classification (i.e., identification) accuracy of 0.78 over a set of 148 users. However, we find that a Random Forest classifier trained on a slightly modified version of this same feature yields an accuracy of 0.93.
摘要
在过去的研究中，键盘动态学已经展示了用户认证的搭配可能性，基于固定文本和自由文本数据。在这项研究中，我们考虑了更加困难的多类用户识别问题，基于自由文本数据。我们尝试使用过去已经用来实现自由文本数据上状态前瞻的复杂图像类特征。使用这种图像类特征和多类卷积神经网络，我们能够获得一个分类精度（即识别率）为0.78，在148个用户中。然而，我们发现，使用一个基于这个特征的微小修改版本的Random Forest分类器，可以达到0.93的精度。
Reinforcement and Deep Reinforcement Learning-based Solutions for Machine Maintenance Planning, Scheduling Policies, and Optimization
For: This paper reviews the applications of reinforcement and deep reinforcement learning for maintenance planning and optimization problems.* Methods: The paper uses a literature review to identify and categorize existing research on reinforcement learning for maintenance planning, and provides graphical and tabular representations of the adopted methodologies, findings, and interpretations.* Results: The paper highlights research gaps, key insights from the literature, and areas for future work in the field of reinforcement learning for maintenance planning.In Simplified Chinese text, the three information points could be summarized as follows:* For: 本文review了使用强化学习和深度强化学习进行维护规划和优化问题的应用。* Methods: 本文使用文献综述来 indentify和分类现有的强化学习维护规划研究，并提供图形和表格形式的采用方法、发现和解释。* Results: 本文指出了维护规划领域的研究漏洞、文献中的关键发现和未来工作的方向。Abstract
Systems and machines undergo various failure modes that result in machine health degradation, so maintenance actions are required to restore them back to a state where they can perform their expected functions. Since maintenance tasks are inevitable, maintenance planning is essential to ensure the smooth operations of the production system and other industries at large. Maintenance planning is a decision-making problem that aims at developing optimum maintenance policies and plans that help reduces maintenance costs, extend asset life, maximize their availability, and ultimately ensure workplace safety. Reinforcement learning is a data-driven decision-making algorithm that has been increasingly applied to develop dynamic maintenance plans while leveraging the continuous information from condition monitoring of the system and machine states. By leveraging the condition monitoring data of systems and machines with reinforcement learning, smart maintenance planners can be developed, which is a precursor to achieving a smart factory. This paper presents a literature review on the applications of reinforcement and deep reinforcement learning for maintenance planning and optimization problems. To capture the common ideas without losing touch with the uniqueness of each publication, taxonomies used to categorize the systems were developed, and reviewed publications were highlighted, classified, and summarized based on these taxonomies. Adopted methodologies, findings, and well-defined interpretations of the reviewed studies were summarized in graphical and tabular representations to maximize the utility of the work for both researchers and practitioners. This work also highlights the research gaps, key insights from the literature, and areas for future work.
摘要
系统和机器会经历多种故障模式，导致机器健康下降，因此维护工作是必要的以还原它们到可以执行预期功能的状态。维护工作是不可避免的，因此维护观念是非常重要，以确保生产系统和其他行业的顺畅运行。维护观念是一个决策问题，旨在发展最佳的维护政策和计划，帮助降低维护成本，延长资产寿命，最大化资产可用性，并确保工作安全。对于维护计划和优化问题，很多使用了强化学习，这是一种基于数据驱动的决策推断算法。通过使用系统和机器的状态监控数据和强化学习，可以开发出智能维护观念，这是一个进攻智能厂的先驱。本文将介绍一篇文献综述，探讨对维护计划和优化问题的应用强化学习和深度强化学习的研究。为了捕捉每篇文献的共同主题而不让它们与具体性的区别，则使用了分类系统，并将综述的文献按照这些分类系统进行分类和摘要。采用的方法、发现和实际的解释都是通过图表和表格的形式呈现，以便对研究人员和实践者具有最大的实用性。本文还强调了研究潜在差距、关键见解和未来工作的方向。
The Ethical Implications of Generative Audio Models: A Systematic Literature Review
results: 研究结果显示，只有少于10%的Generative audio研究文献讨论了可能的负面影响，这是极其罕见的。然而，这些文献中提出的伦理问题和问题是深刻的，例如欺诈、深圳制作和版权侵犯等。本研究这样的缺乏伦理考虑和潜在的负面影响，将是这个领域的未来研究指南。Abstract
Generative audio models typically focus their applications in music and speech generation, with recent models having human-like quality in their audio output. This paper conducts a systematic literature review of 884 papers in the area of generative audio models in order to both quantify the degree to which researchers in the field are considering potential negative impacts and identify the types of ethical implications researchers in this area need to consider. Though 65% of generative audio research papers note positive potential impacts of their work, less than 10% discuss any negative impacts. This jarringly small percentage of papers considering negative impact is particularly worrying because the issues brought to light by the few papers doing so are raising serious ethical implications and concerns relevant to the broader field such as the potential for fraud, deep-fakes, and copyright infringement. By quantifying this lack of ethical consideration in generative audio research and identifying key areas of potential harm, this paper lays the groundwork for future work in the field at a critical point in time in order to guide more conscientious research as this field progresses.
摘要
inTformer: A Time-Embedded Attention-Based Transformer for Crash Likelihood Prediction at Intersections Using Connected Vehicle Data
paper_authors: B. M. Tazbiul Hassan Anik, Zubayer Islam, Mohamed Abdel-Aty
for: 预测交叉点事故可能性的实时预测模型，帮助提高交通安全性。
methods: 使用Transformer模型，通过注意力机制来处理数据序列，并且可以同时处理所有数据元素 durante training。
results: 在使用INRIX和CATT Lab的信号分析平台上测试的connected vehicle数据上，提出了一个名为inTformer的时间嵌入注意力基于Transformer模型，可以效果地预测交叉点事故可能性。最佳inTformer模型达到了73%的敏感性。Abstract
The real-time crash likelihood prediction model is an essential component of the proactive traffic safety management system. Over the years, numerous studies have attempted to construct a crash likelihood prediction model in order to enhance traffic safety, but mostly on freeways. In the majority of the existing studies, researchers have primarily employed a deep learning-based framework to identify crash potential. Lately, Transformer has emerged as a potential deep neural network that fundamentally operates through attention-based mechanisms. Transformer has several functional benefits over extant deep learning models such as Long Short-Term Memory (LSTM), Convolution Neural Network (CNN), etc. Firstly, Transformer can readily handle long-term dependencies in a data sequence. Secondly, Transformers can parallelly process all elements in a data sequence during training. Finally, a Transformer does not have the vanishing gradient issue. Realizing the immense possibility of Transformers, this paper proposes inTersection-Transformer (inTformer), a time-embedded attention-based Transformer model that can effectively predict intersection crash likelihood in real-time. The proposed model was evaluated using connected vehicle data extracted from INRIX and Center for Advanced Transportation Technology (CATT) Lab's Signal Analytics Platform. The data was parallelly formatted and stacked at different timesteps to develop nine inTformer models. The best inTformer model achieved a sensitivity of 73%. This model was also compared to earlier studies on crash likelihood prediction at intersections and with several established deep learning models trained on the same connected vehicle dataset. In every scenario, this inTformer outperformed the benchmark models confirming the viability of the proposed inTformer architecture.
摘要
现实时启发风险预测模型是智能交通安全管理系统的重要组件。过去几年，许多研究都尝试了构建启发风险预测模型，以提高交通安全，但大多数研究都在高速公路上进行。现有的大多数研究者都使用了深度学习框架来识别启发 potential。最近，Transformer 出现了作为深度神经网络的潜在可能，它基于注意力机制来运行。Transformer 与既有的深度学习模型（如 Long Short-Term Memory 和 Convolution Neural Network）相比，具有多种功能优势。首先，Transformer 可以识别长期依赖关系。其次，Transformer 可以并行处理数据序列中的所有元素。最后，Transformer 不受消失梯度问题的影响。鉴于 Transformer 的可能性，本文提出了 intersection-Transformer（inTformer）模型，可以在实时中预测交叉口启发风险。该模型使用 INRIX 和 Center for Advanced Transportation Technology (CATT) Lab 的 Signal Analytics Platform 提供的连接式汽车数据进行评估。数据被平行格式化并堆叠在不同的时间步上，以构建九个 inTformer 模型。最佳 inTformer 模型达到了 73% 的敏感度。这个模型还与其他关于交叉口启发风险预测的研究和已有的深度学习模型在同一个连接式汽车dataset上进行比较。在每种场景下，这个 inTformer 模型都超越了参考模型，证明了提案的 inTformer 架构的可行性。
Optimal Learners for Realizable Regression: PAC Learning and Online Learning
results: 本文确定了一个必要条件 для学习可能性，并 conjecture 这可能也是充分条件。此外，本文还解决了 Daskalakis 和 Golowich 在 STOC ‘22 中提出的一个开问。Abstract
In this work, we aim to characterize the statistical complexity of realizable regression both in the PAC learning setting and the online learning setting. Previous work had established the sufficiency of finiteness of the fat shattering dimension for PAC learnability and the necessity of finiteness of the scaled Natarajan dimension, but little progress had been made towards a more complete characterization since the work of Simon 1997 (SICOMP '97). To this end, we first introduce a minimax instance optimal learner for realizable regression and propose a novel dimension that both qualitatively and quantitatively characterizes which classes of real-valued predictors are learnable. We then identify a combinatorial dimension related to the Graph dimension that characterizes ERM learnability in the realizable setting. Finally, we establish a necessary condition for learnability based on a combinatorial dimension related to the DS dimension, and conjecture that it may also be sufficient in this context. Additionally, in the context of online learning we provide a dimension that characterizes the minimax instance optimal cumulative loss up to a constant factor and design an optimal online learner for realizable regression, thus resolving an open question raised by Daskalakis and Golowich in STOC '22.
摘要
在这项工作中，我们目标是Characterize realizable regression的统计复杂性在PAC学习设定下和在线学习设定下。previoius work had established the sufficiency of finiteness of the fat shattering dimension for PAC learnability and the necessity of finiteness of the scaled Natarajan dimension, but little progress had been made towards a more complete characterization since the work of Simon 1997 (SICOMP '97). To this end, we first introduce a minimax instance optimal learner for realizable regression and propose a novel dimension that both qualitatively and quantitatively characterizes which classes of real-valued predictors are learnable. We then identify a combinatorial dimension related to the Graph dimension that characterizes ERM learnability in the realizable setting. Finally, we establish a necessary condition for learnability based on a combinatorial dimension related to the DS dimension, and conjecture that it may also be sufficient in this context. 在线学习上，我们提供一个characterizes the minimax instance optimal cumulative loss up to a constant factor的dimension，并设计了一个optimal online learner for realizable regression，thereby resolving an open question raised by Daskalakis and Golowich in STOC '22.
RADAR: Robust AI-Text Detection via Adversarial Learning
results: 对8种LLMs（Pythia、Dolly 2.0、Palmyra、Camel、GPT-J、Dolly 1.0、LLaMA、Vicuna）在4个dataset上进行了实验，结果显示，RADARsignificantly outperforms现有的AI-文本检测方法，特别是在paraphrasing存在时。此外，我们还发现RADAR在instruction-tuned LLMs上 Transferability 强，并且通过GPT-3.5进行评估，发现RADAR的改进能力。Abstract
Recent advances in large language models (LLMs) and the intensifying popularity of ChatGPT-like applications have blurred the boundary of high-quality text generation between humans and machines. However, in addition to the anticipated revolutionary changes to our technology and society, the difficulty of distinguishing LLM-generated texts (AI-text) from human-generated texts poses new challenges of misuse and fairness, such as fake content generation, plagiarism, and false accusation of innocent writers. While existing works show that current AI-text detectors are not robust to LLM-based paraphrasing, this paper aims to bridge this gap by proposing a new framework called RADAR, which jointly trains a Robust AI-text Detector via Adversarial leaRning. RADAR is based on adversarial training of a paraphraser and a detector. The paraphraser's goal is to generate realistic contents to evade AI-text detection. RADAR uses the feedback from the detector to update the paraphraser, and vice versa. Evaluated with 8 different LLMs (Pythia, Dolly 2.0, Palmyra, Camel, GPT-J, Dolly 1.0, LLaMA, and Vicuna) across 4 datasets, experimental results show that RADAR significantly outperforms existing AI-text detection methods, especially when paraphrasing is in place. We also identify the strong transferability of RADAR from instruction-tuned LLMs to other LLMs, and evaluate the improved capability of RADAR via GPT-3.5.
摘要
RADAR is based on adversarial training of a paraphraser and a detector. The paraphraser's goal is to generate realistic contents to evade AI-text detection, while the detector's goal is to correctly identify AI-generated texts. RADAR uses the feedback from the detector to update the paraphraser, and vice versa. The framework was evaluated with 8 different LLMs (Pythia, Dolly 2.0, Palmyra, Camel, GPT-J, Dolly 1.0, LLaMA, and Vicuna) across 4 datasets, and the results show that RADAR significantly outperforms existing AI-text detection methods, especially when paraphrasing is involved. Additionally, RADAR was found to have strong transferability from instruction-tuned LLMs to other LLMs, and its capability was improved further via GPT-3.5.
Effect of Intensity Standardization on Deep Learning for WML Segmentation in Multi-Centre FLAIR MRI
results: 结果表明，IAMLAB和Ensemble方法在不同的lesion category中均有更高的WML segmentation性能，比如原始数据或其他normalization方法。IAMLAB和Ensemble方法在各种lesion category中都有最高的dice similarity coefficient（DSC），并且在不同的临床数据集中也具有最高的DSC。这些方法可以减轻MRI领域的差异，并且是适用于DL-based WML segmentation的优选方法。Abstract
Deep learning (DL) methods for white matter lesion (WML) segmentation in MRI suffer a reduction in performance when applied on data from a scanner or centre that is out-of-distribution (OOD) from the training data. This is critical for translation and widescale adoption, since current models cannot be readily applied to data from new institutions. In this work, we evaluate several intensity standardization methods for MRI as a preprocessing step for WML segmentation in multi-centre Fluid-Attenuated Inversion Recovery (FLAIR) MRI. We evaluate a method specifically developed for FLAIR MRI called IAMLAB along with other popular normalization techniques such as White-strip, Nyul and Z-score. We proposed an Ensemble model that combines predictions from each of these models. A skip-connection UNet (SC UNet) was trained on the standardized images, as well as the original data and segmentation performance was evaluated over several dimensions. The training (in-distribution) data consists of a single study, of 60 volumes, and the test (OOD) data is 128 unseen volumes from three clinical cohorts. Results show IAMLAB and Ensemble provide higher WML segmentation performance compared to models from original data or other normalization methods. IAMLAB & Ensemble have the highest dice similarity coefficient (DSC) on the in-distribution data (0.78 & 0.80) and on clinical OOD data. DSC was significantly higher for IAMLAB compared to the original data (p<0.05) for all lesion categories (LL>25mL: 0.77 vs. 0.71; 10mL<= LL<25mL: 0.66 vs. 0.61; LL<10mL: 0.53 vs. 0.52). The IAMLAB and Ensemble normalization methods are mitigating MRI domain shift and are optimal for DL-based WML segmentation in unseen FLAIR data.
摘要
深度学习（DL）方法 для白 matter损害（WML）分割在MRI中受到外部数据集（OOD）的影响，导致性能下降。这对于翻译和大规模应用而言是关键，因为现有的模型无法直接应用于新机构的数据集。在这项工作中，我们评估了多种MRIIntensity标准化方法作为WML分割前的预处理步骤。我们评估了特定 дляFLAIR MRI的IAMLAB方法，以及其他流行的标准化技术，如白带、恩琴和Z-score。我们提出了一种组合这些模型的ensemble模型。一个skip-connection UNet（SC UNNet）在标准化图像上进行训练，以及原始数据上进行分割性能的评估。训练（卷积）数据集包括60个Volume，测试（OOD）数据集包括128个未看过的Volume从三个临床各类数据集。结果显示，IAMLAB和Ensemble模型在WML分割性能方面比原始数据或其他标准化方法高得多。IAMLAB和Ensemble模型在卷积数据集（卷积数据）上的DSC值分别为0.78和0.80，并在临床OOD数据上达到了最高的DSC值（0.77和0.80）。对于所有损害类别（LL>25mL：0.77 vs. 0.71；10mL≤ LL<25mL：0.66 vs. 0.61；LL<10mL：0.53 vs. 0.52），IAMLAB模型的DSC值与原始数据相比有 statistically significant difference（P<0.05）。IAMLAB和Ensemble normalization方法可以 Mitigate MRI域shift，是适用于DL基于WML分割的优选方案。
A Combinatorial Characterization of Online Learning Games with Bounded Losses
results: 在vector-valued regression和多标签分类两个自然的学习设定中，得到了第一个量化的在线学习性能CharacterizationAbstract
We study the online learnability of hypothesis classes with respect to arbitrary, but bounded, loss functions. We give a new scale-sensitive combinatorial dimension, named the sequential Minimax dimension, and show that it gives a tight quantitative characterization of online learnability. As applications, we give the first quantitative characterization of online learnability for two natural learning settings: vector-valued regression and multilabel classification.
摘要
我们研究在使用各种固定但受限的损失函数时，假设集合在线学习的可学习性。我们提出了一种新的敏感度量，称为顺序最小最大维度，并证明它为在线学习的准确量提供了紧跟的量化特征化。我们还应用到了两个自然的学习场景：向量值回归和多类分类。
Controlling Chaotic Maps using Next-Generation Reservoir Computing
results: 论文在一系列控制任务中展示了控制器的性能，包括控制系统 между不稳定的固定点、稳定系统到更高阶 периодических轨迹、和到一个指定的状态。 论文还表明了控制器只需要10个数据点进行训练，可以在一次迭代中控制系统到指定的轨迹，并且对噪音和模型误差 Displaytext 有 robustness。Abstract
In this work, we combine nonlinear system control techniques with next-generation reservoir computing, a best-in-class machine learning approach for predicting the behavior of dynamical systems. We demonstrate the performance of the controller in a series of control tasks for the chaotic H\'enon map, including controlling the system between unstable fixed-points, stabilizing the system to higher order periodic orbits, and to an arbitrary desired state. We show that our controller succeeds in these tasks, requires only 10 data points for training, can control the system to a desired trajectory in a single iteration, and is robust to noise and modeling error.
摘要
在这项工作中，我们将非线性系统控制技术与下一代散射 computing（一种最佳级机器学习方法）结合使用，用于预测动力系统的行为。我们在哈农地图中进行了一系列控制任务，包括控制系统在不稳定的固定点上，稳定系统到更高阶 periodic orbit 以及到任意所希望的状态。我们发现，我们的控制器在这些任务中具有出色的表现，只需要10个数据点进行训练，可以在单次迭代中控制系统到所希望的轨迹，并具有噪声和模型误差的抗性。
For Women, Life, Freedom: A Participatory AI-Based Social Web Analysis of a Watershed Moment in Iran’s Gender Struggles
paper_authors: Adel Khorramrouz, Sujan Dutta, Ashiqur R. KhudaBukhsh
for: 这 paper 的目的是计算并分析波斯语推特讨论，以估算在警察拘留中死亡的马哈萨·阿米尼去世后，对男女平等的态度发生了如何的变化。
methods: 该 paper 使用了一个ensemble active learning挺训练一个立场分类器，其特点在于伊朗女性参与了活动的角色，不仅提供标签，还提供了有价值的关键词 для更加有意义的词汇创造以及短示文档 для导向采样步骤。
results: 分析结果表明，贝娅·阿米尼去世后，波斯语推特讨论发生了偏好化的变化，双方的负面和正面的推特数量均增加了，其中正面推特数量微小地大于负面推特数量增加。此外，与基eline波斯推特活动相比，支持抗议的推特账户的创建时间更加接近于基eline。Abstract
In this paper, we present a computational analysis of the Persian language Twitter discourse with the aim to estimate the shift in stance toward gender equality following the death of Mahsa Amini in police custody. We present an ensemble active learning pipeline to train a stance classifier. Our novelty lies in the involvement of Iranian women in an active role as annotators in building this AI system. Our annotators not only provide labels, but they also suggest valuable keywords for more meaningful corpus creation as well as provide short example documents for a guided sampling step. Our analyses indicate that Mahsa Amini's death triggered polarized Persian language discourse where both fractions of negative and positive tweets toward gender equality increased. The increase in positive tweets was slightly greater than the increase in negative tweets. We also observe that with respect to account creation time, between the state-aligned Twitter accounts and pro-protest Twitter accounts, pro-protest accounts are more similar to baseline Persian Twitter activity.
摘要
在这篇论文中，我们对波斯语推特讨论进行计算分析，以估算死于警察执法中的马赛穆罕默的去世对性别平等的看法产生影响。我们提出了一个协同学习激活管道，以训练立场分类器。我们的创新在于伊朗女性参与了活动角色，作为标注人员，不仅提供标签，还提供了有价值的关键词，以便更好地创建词汇库，以及短示例文档，用于指导采样步骤。我们的分析表明，马赛穆罕默的去世导致波斯语推特讨论呈极化趋势，负面和正面的推特数量均增加，但正面推特数量略大于负面推特数量。此外，我们发现，在账户创建时间方面，支持抗议的推特账户和支持政府的推特账户之间的差异较小。
Predicting Outcomes in Long COVID Patients with Spatiotemporal Attention
results: 在具有困难预测特征的长COVID患者数据集上，本方法比相关方法表现出色，可用于评估长COVID患者的严重程度。Abstract
Long COVID is a general term of post-acute sequelae of COVID-19. Patients with long COVID can endure long-lasting symptoms including fatigue, headache, dyspnea and anosmia, etc. Identifying the cohorts with severe long-term complications in COVID-19 could benefit the treatment planning and resource arrangement. However, due to the heterogeneous phenotype presented in long COVID patients, it is difficult to predict their outcomes from their longitudinal data. In this study, we proposed a spatiotemporal attention mechanism to weigh feature importance jointly from the temporal dimension and feature space. Considering that medical examinations can have interchangeable orders in adjacent time points, we restricted the learning of short-term dependency with a Local-LSTM and the learning of long-term dependency with the joint spatiotemporal attention. We also compared the proposed method with several state-of-the-art methods and a method in clinical practice. The methods are evaluated on a hard-to-acquire clinical dataset of patients with long COVID. Experimental results show the Local-LSTM with joint spatiotemporal attention outperformed related methods in outcome prediction. The proposed method provides a clinical tool for the severity assessment of long COVID.
摘要
长期 COVID 是 COVID-19 后遗症的总称。患有长期 COVID 的患者可能会经历长期的症状，如疲劳、头痛、呼吸急促和 anosmia 等。确定 COVID-19 患者长期grave的合并症状可以帮助诊断和资源安排。然而，由于长期 COVID 患者的多样性表现，从 их长期数据来预测结果很困难。在这项研究中，我们提出了一种空间时间注意力机制，以同时评估特征的重要性。由于医学检查可能会在邻近时间点互换检查顺序，我们限制了短期依赖性学习的 Local-LSTM，以及与其他特征空间进行同时学习的合并空间时间注意力。我们还对我们的方法与其他现有方法和临床实践中的方法进行了比较。这些方法在普遍难以获得的医学数据集上进行了评估。实验结果表明，Local-LSTM WITH 共同空间时间注意力在结果预测方面表现出色，超过了相关方法。我们的方法提供了诊断长期 COVID 严重程度的临床工具。
Formulation Graphs for Mapping Structure-Composition of Battery Electrolytes to Device Performance
paper_authors: Vidushi Sharma, Maxwell Giammona, Dmitry Zubarev, Andy Tek, Khanh Nugyuen, Linda Sundberg, Daniele Congiu, Young-Hye La
For: The paper is written for researchers and developers working on the discovery and development of new combinatorial materials, particularly in the context of battery electrolytes.* Methods: The paper proposes a deep learning model called Formulation Graph Convolution Network (F-GCN) that can predict the properties of liquid formulations based on the structure-composition relationship of their individual components. The model uses molecular descriptors derived from molecular graphs, informed by HOMO-LUMO and electric moment properties of the molecules.* Results: The paper demonstrates the effectiveness of the proposed model on two exemplary datasets related to battery electrolytes, achieving low errors in predicting performance metrics such as Coulombic Efficiency (CE) and specific capacity. The best performing F-GCN model uses molecular descriptors derived from molecular graphs that are informed with HOMO-LUMO and electric moment properties of the molecules.Abstract
Advanced computational methods are being actively sought for addressing the challenges associated with discovery and development of new combinatorial material such as formulations. A widely adopted approach involves domain informed high-throughput screening of individual components that can be combined into a formulation. This manages to accelerate the discovery of new compounds for a target application but still leave the process of identifying the right 'formulation' from the shortlisted chemical space largely a laboratory experiment-driven process. We report a deep learning model, Formulation Graph Convolution Network (F-GCN), that can map structure-composition relationship of the individual components to the property of liquid formulation as whole. Multiple GCNs are assembled in parallel that featurize formulation constituents domain-intuitively on the fly. The resulting molecular descriptors are scaled based on respective constituent's molar percentage in the formulation, followed by formalizing into a combined descriptor that represents a complete formulation to an external learning architecture. The use case of proposed formulation learning model is demonstrated for battery electrolytes by training and testing it on two exemplary datasets representing electrolyte formulations vs battery performance -- one dataset is sourced from literature about Li/Cu half-cells, while the other is obtained by lab-experiments related to lithium-iodide full-cell chemistry. The model is shown to predict the performance metrics like Coulombic Efficiency (CE) and specific capacity of new electrolyte formulations with lowest reported errors. The best performing F-GCN model uses molecular descriptors derived from molecular graphs that are informed with HOMO-LUMO and electric moment properties of the molecules using a knowledge transfer technique.
摘要
当前计算方法在开发新的 combinatorial材料领域中是活跃的搜寻。一种广泛采用的方法是通过域内高速屏选个别组分，以加速针对特定应用的新化合物的发现。然而，从短列表中选择合适的“形态”仍然是实验室实验驱动的过程。我们报道了一种深度学习模型，形态图 convolutional neural network (F-GCN)，可以将个体组分的结构-组分关系映射到液体形态的性能。多个GCN被紧密地 assembled 并在 fly 上域特征化形态成分。 resulting molecular descriptors 被权重根据各个成分的分子比例缩放，然后以组合的描述符形式传递给外部学习架构。我们使用了 Li/Cu 半细胞和锂iodide 全细胞化学的两个数据集来评估我们的形态学习模型。我们的模型可以预测新的电解质形态的性能指标，如电子效率（CE）和Specific capacity。我们的最佳运行F-GCN模型使用基于分子图的分子描述符，并使用了知识传递技术以获得HOMO-LUMO和电动量特性。
URL: A Representation Learning Benchmark for Transferable Uncertainty Estimates
results: 研究发现，对于表示本身的不确定性或直接估计预测风险的方法比较出色，但是实现可传输的不确定量衡仍然是一个开放的挑战。Abstract
Representation learning has significantly driven the field to develop pretrained models that can act as a valuable starting point when transferring to new datasets. With the rising demand for reliable machine learning and uncertainty quantification, there is a need for pretrained models that not only provide embeddings but also transferable uncertainty estimates. To guide the development of such models, we propose the Uncertainty-aware Representation Learning (URL) benchmark. Besides the transferability of the representations, it also measures the zero-shot transferability of the uncertainty estimate using a novel metric. We apply URL to evaluate eleven uncertainty quantifiers that are pretrained on ImageNet and transferred to eight downstream datasets. We find that approaches that focus on the uncertainty of the representation itself or estimate the prediction risk directly outperform those that are based on the probabilities of upstream classes. Yet, achieving transferable uncertainty quantification remains an open challenge. Our findings indicate that it is not necessarily in conflict with traditional representation learning goals. Code is provided under https://github.com/mkirchhof/url .
摘要
“表达学学习”在领域中发挥了重要作用，使得开发者可以从新数据集上转移到新的任务。随着机器学习的可靠性和不确定性评估的需求增加，需要开发可以提供嵌入和传输不确定性估计的预训练模型。为了引导这类模型的开发，我们提出了“不确定性感知学习”（URL）数据集。除了表达的传输性外，它还测量了零批转移不确定性估计的新指标。我们对ImageNet预训练的十一种不确定量进行了URL的评估，并将其转移到八个下游数据集上。我们发现，关注表达不确定性本身或直接估计预测风险的方法表现较好。然而，实现传输不确定性估计仍然是一个开放的挑战。我们的发现表明，这并不是传统表达学习目标的矛盾。代码可以在获取。
A Theoretical Perspective on Subnetwork Contributions to Adversarial Robustness
results: 该论文提出了一个新的理论框架，用于研究攻击如何影响整个网络的Robustness，并提供了一种测试这种理论的方法。经验表明，如果某个子网络具有一定的鲁棒性，那么整个网络也是鲁棒的，并且需要在不同层次之间存在一定的依赖关系。Abstract
The robustness of deep neural networks (DNNs) against adversarial attacks has been studied extensively in hopes of both better understanding how deep learning models converge and in order to ensure the security of these models in safety-critical applications. Adversarial training is one approach to strengthening DNNs against adversarial attacks, and has been shown to offer a means for doing so at the cost of applying computationally expensive training methods to the entire model. To better understand these attacks and facilitate more efficient adversarial training, in this paper we develop a novel theoretical framework that investigates how the adversarial robustness of a subnetwork contributes to the robustness of the entire network. To do so we first introduce the concept of semirobustness, which is a measure of the adversarial robustness of a subnetwork. Building on this concept, we then provide a theoretical analysis to show that if a subnetwork is semirobust and there is a sufficient dependency between it and each subsequent layer in the network, then the remaining layers are also guaranteed to be robust. We validate these findings empirically across multiple DNN architectures, datasets, and adversarial attacks. Experiments show the ability of a robust subnetwork to promote full-network robustness, and investigate the layer-wise dependencies required for this full-network robustness to be achieved.
摘要
深度神经网络（DNNs）的对抗攻击的稳定性已经得到了广泛的研究，以便更好地理解深度学习模型的协调方式，并确保这些模型在安全关键应用中的安全性。对抗训练是一种加强DNNs对抗攻击的方法，并已经证明可以通过对整个模型进行计算昂贵的训练方法来实现。为了更好地理解这些攻击和实现更有效的对抗训练，在这篇论文中我们开发了一种新的理论框架，以 investigate如何各个子网络的对抗稳定性对整个网络的稳定性的贡献。我们首先介绍了semirobustness这个概念，它是一种对抗稳定性的度量。然后，我们提供了一种理论分析，证明如果一个子网络是semirobust的，并且每个后续层与其之间存在足够的依赖关系，那么剩下的层也一定是稳定的。我们验证了这些发现的实验结果，并对多个DNN架构、数据集和攻击方法进行了验证。实验表明，一个稳定的子网络可以推动整个网络的稳定性，并且调查层间的依赖关系可以实现这种全网络稳定性。
CLIPMasterPrints: Fooling Contrastive Language-Image Pre-training Using Latent Variable Evolution
results: 研究发现，使用少量图像标签可以生成大量semantically相关的图像，而且这些图像可以让 CLIP 模型具有高置信度。此外，研究还发现 modality gap 在多modal网络中导致 CLIP 模型易受到伪装主要图像的攻击。Abstract
Models leveraging both visual and textual data such as Contrastive Language-Image Pre-training (CLIP), are increasingly gaining importance. In this work, we show that despite their versatility, such models are vulnerable to what we refer to as fooling master images. Fooling master images are capable of maximizing the confidence score of a CLIP model for a significant number of widely varying prompts, while being unrecognizable for humans. We demonstrate how fooling master images can be mined by searching the latent space of generative models by means of an evolution strategy or stochastic gradient descent. We investigate the properties of the mined fooling master images, and find that images trained on a small number of image captions potentially generalize to a much larger number of semantically related captions. Further, we evaluate two possible mitigation strategies and find that vulnerability to fooling master examples is closely related to a modality gap in contrastive pre-trained multi-modal networks. From the perspective of vulnerability to off-manifold attacks, we therefore argue for the mitigation of modality gaps in CLIP and related multi-modal approaches. Source code and mined CLIPMasterPrints are available at https://github.com/matfrei/CLIPMasterPrints.
摘要
模型利用视觉和文本数据，如对照语言图像预训练（CLIP），在当前研究中变得越来越重要。在这项工作中，我们表明，尽管这些模型具有多样性，但它们却容易受到我们称为“欺骗主图”的攻击。欺骗主图可以让CLIP模型对各种各样的提示进行最大化信任分数，而human是无法识别的。我们表明，可以通过演化策略或权重下降来在生成模型的latent空间中搜寻欺骗主图。我们研究欺骗主图的性质，发现图像通过少量的图像描述训练可以对Semantically相关的提示进行扩展。此外，我们评估了两种可能的防范策略，发现攻击模式与多样性差有close关系。从防范多样性差的角度来看，我们 argue for the mitigation of modality gaps in CLIP and related multi-modal approaches。代码和搜寻到的CLIPMasterPrints可以在https://github.com/matfrei/CLIPMasterPrints上获取。
Exploring the Lottery Ticket Hypothesis with Explainability Methods: Insights into Sparse Network Performance
results: 研究发现，随着权重的减少，网络的性能逐渐下降，并且发现了原始网络中的概念和像素与减少后的网络存在差异，这可能是性能下降的原因。Abstract
Discovering a high-performing sparse network within a massive neural network is advantageous for deploying them on devices with limited storage, such as mobile phones. Additionally, model explainability is essential to fostering trust in AI. The Lottery Ticket Hypothesis (LTH) finds a network within a deep network with comparable or superior performance to the original model. However, limited study has been conducted on the success or failure of LTH in terms of explainability. In this work, we examine why the performance of the pruned networks gradually increases or decreases. Using Grad-CAM and Post-hoc concept bottleneck models (PCBMs), respectively, we investigate the explainability of pruned networks in terms of pixels and high-level concepts. We perform extensive experiments across vision and medical imaging datasets. As more weights are pruned, the performance of the network degrades. The discovered concepts and pixels from the pruned networks are inconsistent with the original network -- a possible reason for the drop in performance.
摘要
发现高性能的简化网络在巨量神经网络中是有利于在具有有限存储的设备上部署，如移动电话。此外，AI信任的重要因素之一是模型解释性。抽奖假设（LTH）找到了深度网络中的相似或更高性能的网络，但有限的研究对LTH的成功或失败进行了解释性研究。在这项工作中，我们研究了剪除网络性能的提高或下降的原因。使用Grad-CAM和后置概念瓶颈模型（PCBM），我们进行了剪除网络的解释性研究，即像素和高级概念的解释性。我们在视觉和医学影像 dataset 上进行了广泛的实验。随着更多的权重被剪除，网络性能下降。发现的概念和像素从剪除网络中与原始网络不同，可能是性能下降的原因。
Neural Abstraction-Based Controller Synthesis and Deployment
results: 我们的方法可以减少抽象基本方法的内存占用，并且可以在实时控制中应用。在选择的标准套件中，我们的方法可以减少平均内存占用量分别为1.31×10^5和7.13×10^3，最高达7.54×10^5和3.18×10^4。Abstract
Abstraction-based techniques are an attractive approach for synthesizing correct-by-construction controllers to satisfy high-level temporal requirements. A main bottleneck for successful application of these techniques is the memory requirement, both during controller synthesis and in controller deployment. We propose memory-efficient methods for mitigating the high memory demands of the abstraction-based techniques using neural network representations. To perform synthesis for reach-avoid specifications, we propose an on-the-fly algorithm that relies on compressed neural network representations of the forward and backward dynamics of the system. In contrast to usual applications of neural representations, our technique maintains soundness of the end-to-end process. To ensure this, we correct the output of the trained neural network such that the corrected output representations are sound with respect to the finite abstraction. For deployment, we provide a novel training algorithm to find a neural network representation of the synthesized controller and experimentally show that the controller can be correctly represented as a combination of a neural network and a look-up table that requires a substantially smaller memory. We demonstrate experimentally that our approach significantly reduces the memory requirements of abstraction-based methods. For the selected benchmarks, our approach reduces the memory requirements respectively for the synthesis and deployment by a factor of $1.31\times 10^5$ and $7.13\times 10^3$ on average, and up to $7.54\times 10^5$ and $3.18\times 10^4$. Although this reduction is at the cost of increased off-line computations to train the neural networks, all the steps of our approach are parallelizable and can be implemented on machines with higher number of processing units to reduce the required computational time.
摘要
“对于高水平时间需求的正确控制器的合成，具有吸引力的方法是基于抽象的技术。然而，这些技术的记忆需求在控制器合成和部署过程中都是主要的瓶颈。我们提出了一些记忆效率的方法，使用神经网络表示法来减少高级抽象技术的记忆需求。为了实现这些目的，我们提出了一种在线算法，它基于压缩神经网络表示法来进行控制器合成。与传统神经网络应用不同的是，我们的方法保持了途径的有效性。为了确保这一点，我们会对训练神经网络输出进行修正，使其与有限抽象之间保持相对的准确性。在部署阶段，我们提供了一种新的训练算法，用于在神经网络和lookup表之间找到一个可以减少记忆需求的控制器表示。我们通过实验证明，我们的方法可以减少抽象基于方法的记忆需求。对于我们选择的标准套件，我们的方法分别减少了在合成和部署阶段的记忆需求的平均值为1.31\*10^5和7.13\*10^3。最多可以减少到7.54\*10^5和3.18\*10^4。虽然这些减少是在训练神经网络的过程中付出的成本，但所有的步骤都可以并行进行并在高处理器数量的机器上进行实现，以减少所需的计算时间。”
results: 本文提出了一些对ID算法的批评和改进，包括指出ID算法在某些情况下会失败，并提供了一些图形化的Characterization来描述这些情况。Abstract
The ID algorithm solves the problem of identification of interventional distributions of the form p(Y | do(a)) in graphical causal models, and has been formulated in a number of ways [12, 9, 6]. The ID algorithm is sound (outputs the correct functional of the observed data distribution whenever p(Y | do(a)) is identified in the causal model represented by the input graph), and complete (explicitly flags as a failure any input p(Y | do(a)) whenever this distribution is not identified in the causal model represented by the input graph). The reference [9] provides a result, the so called "hedge criterion" (Corollary 3), which aims to give a graphical characterization of situations when the ID algorithm fails to identify its input in terms of a structure in the input graph called the hedge. While the ID algorithm is, indeed, a sound and complete algorithm, and the hedge structure does arise whenever the input distribution is not identified, Corollary 3 presented in [9] is incorrect as stated. In this note, I outline the modern presentation of the ID algorithm, discuss a simple counterexample to Corollary 3, and provide a number of graphical characterizations of the ID algorithm failing to identify its input distribution.
摘要
“ID算法解决了图structural causal模型中p(Y | do(a))的分布标定问题，并在多种形式下表述（[12, 9, 6]）。ID算法是有效的（对于输入分布p(Y | do(a))，输出正确的函数），并且是完整的（如果输入分布不能在图structural causal模型中标定，则直接标记为失败）。”“参考[9]中的结论（即‘别branch criterion’）提供了一种图Structural characterization of situations when the ID algorithm fails to identify its input, in terms of a structure in the input graph called the hedge. However, this conclusion is incorrect as stated, and in this note, I present a modern presentation of the ID algorithm and a simple counterexample to Corollary 3. Additionally, I provide several graphical characterizations of the ID algorithm failing to identify its input distribution.”
Incentive-Theoretic Bayesian Inference for Collaborative Science
results: 论文表明了执行者可以通过代理人的决策行为来揭示部分信息，并使用这些信息来控制 posterior 概率的值。这一结果有一个重要的应用是在临床试验中设置类型一错误水平：试验类型一错误水平应该是临床试验成本除以企业利润的比率。Abstract
Contemporary scientific research is a distributed, collaborative endeavor, carried out by teams of researchers, regulatory institutions, funding agencies, commercial partners, and scientific bodies, all interacting with each other and facing different incentives. To maintain scientific rigor, statistical methods should acknowledge this state of affairs. To this end, we study hypothesis testing when there is an agent (e.g., a researcher or a pharmaceutical company) with a private prior about an unknown parameter and a principal (e.g., a policymaker or regulator) who wishes to make decisions based on the parameter value. The agent chooses whether to run a statistical trial based on their private prior and then the result of the trial is used by the principal to reach a decision. We show how the principal can conduct statistical inference that leverages the information that is revealed by an agent's strategic behavior -- their choice to run a trial or not. In particular, we show how the principal can design a policy to elucidate partial information about the agent's private prior beliefs and use this to control the posterior probability of the null. One implication is a simple guideline for the choice of significance threshold in clinical trials: the type-I error level should be set to be strictly less than the cost of the trial divided by the firm's profit if the trial is successful.
摘要
现代科学研究是一项分布式、合作性的努力，由研究人员、管理机构、资金机构、商业伙伴和科学机构共同参与，这些组织之间存在不同的驱动和激励。为保持科学的严谨性，统计方法应该考虑这种情况。为此，我们研究在有一个代理人（例如研究人员或药品制造商）有私人估计参数的情况下，检测 гипотезы的问题。代理人根据自己的私人估计选择是否进行统计试验，然后试验结果被使用者来做决策。我们表明如何使得首脑可以通过代理人的战略行为（即是否进行试验）了解一部分私人估计信息，并使用这些信息来控制后验 posterior 概率。特别是，我们表明如何使得首脑可以设计一种政策来描述代理人的私人估计信息，并使用这些信息来控制后验 posterior 概率。这一结论之一是一个简单的临床试验选择水平指南：类型一错误水平应该设置为试验成本除以成功后利润的比率。
QIGen: Generating Efficient Kernels for Quantized Inference on Large Language Models
paper_authors: Tommaso Pegolotti, Elias Frantar, Dan Alistarh, Markus Püschel
for: 支持量化生成推理在 LLMA 或 OPT 上的自动代码生成方法。
methods: 基于目标架构和性能模型，包括硬件特性和方法特有的准确性约束。
results: CPU 上的 LLMA 模型推理显示，我们的方法可以达到高性能和高准确性，与现有开源解决方案相比。Here’s the English version for reference:
for: An automatic code generation approach for supporting quantized generative inference on large language models (LLMs) such as LLaMA or OPT on off-the-shelf CPUs.
methods: Informed by the target architecture and a performance model, including both hardware characteristics and method-specific accuracy constraints.
results: Results on CPU-based inference for LLaMA models show that our approach can lead to high performance and high accuracy, comparing favorably to the best existing open-source solution.Abstract
We present ongoing work on a new automatic code generation approach for supporting quantized generative inference on LLMs such as LLaMA or OPT on off-the-shelf CPUs. Our approach is informed by the target architecture and a performance model, including both hardware characteristics and method-specific accuracy constraints. Results on CPU-based inference for LLaMA models show that our approach can lead to high performance and high accuracy, comparing favorably to the best existing open-source solution. A preliminary implementation is available at https://github.com/IST-DASLab/QIGen.
摘要
我们正在进行一项新的自动代码生成方法，用于支持量化生成推理在LLaMA或OPT类型的语言模型上的Off-the-shelf CPU上。我们的方法受到目标架构和性能模型的影响，包括硬件特性和方法具体精度限制。对CPU上的推理过程中的LLaMA模型进行了结果，我们的方法可以实现高性能和高精度，与现有开源解决方案相比，表现优异。一个初步的实现可以在https://github.com/IST-DASLab/QIGen上获得。
Polybot: Training One Policy Across Robots While Embracing Variability
paper_authors: Jonathan Yang, Dorsa Sadigh, Chelsea Finn for:多种机器人平台上的视觉控制技能的跨平台传输methods:使用终端摄像头和团队代码库实现观察和动作空间的匹配使用对比学习对策的内部表示进行匹配results:在6个任务和3个机器人上收集了60小时的数据，并取得了显著提高成功率和样本效率的结果，验证了我们的设计决策。Please note that the above information is in Simplified Chinese text, as requested.Abstract
Reusing large datasets is crucial to scale vision-based robotic manipulators to everyday scenarios due to the high cost of collecting robotic datasets. However, robotic platforms possess varying control schemes, camera viewpoints, kinematic configurations, and end-effector morphologies, posing significant challenges when transferring manipulation skills from one platform to another. To tackle this problem, we propose a set of key design decisions to train a single policy for deployment on multiple robotic platforms. Our framework first aligns the observation and action spaces of our policy across embodiments via utilizing wrist cameras and a unified, but modular codebase. To bridge the remaining domain shift, we align our policy's internal representations across embodiments through contrastive learning. We evaluate our method on a dataset collected over 60 hours spanning 6 tasks and 3 robots with varying joint configurations and sizes: the WidowX 250S, the Franka Emika Panda, and the Sawyer. Our results demonstrate significant improvements in success rate and sample efficiency for our policy when using new task data collected on a different robot, validating our proposed design decisions. More details and videos can be found on our anonymized project website: https://sites.google.com/view/polybot-multirobot
摘要
重用大量数据是扩展视觉基于机器人 manipulate 到日常场景中的关键因素，因为收集机器人数据的成本很高。然而，机器人平台具有不同的控制方案、摄像头视点、骨骼配置和器官形态，这会导致将抓取技能从一个平台转移到另一个平台的具有很大挑战。为解决这个问题，我们提出了一些关键的设计决策，用于在多个机器人平台上训练单个策略。我们的框架首先将我们策略的观察空间和动作空间在不同实现中进行对齐，通过使用臂部摄像头和一个统一、可分模块的代码库来实现这一点。为了填补剩下的领域差异，我们对策略的内部表示进行对齐，通过对比学习来实现这一点。我们的方法在一个包含60小时、6个任务和3个机器人的数据集上进行了评估，这些机器人包括WidowX 250S、Franka Emika Panda 和Sawyer。我们的结果表明，使用我们的策略在新的任务数据上进行训练后，在不同机器人上的成功率和样本效率有显著提高，这 validate 我们的设计决策。更多细节和视频可以在我们匿名项目网站上找到：https://sites.google.com/view/polybot-multirobot。
SAR: Generalization of Physiological Agility and Dexterity via Synergistic Action Representation
results: 结果表明，使用SAR在更复杂任务中能够显著提高学习效率和成功率。在人工手和脚模型中，SAR-使用策略可以在各种 Terrains上实现 robust locomotion，而基线方法无法学习有意义的行为。此外，在多个目标搅拌任务中，SAR-使用策略的成功率高达70%，而基线方法的成功率仅有20%。这两个SAR-使用策略还能够在不同的环境条件下进行零基础学习，而不使用SAR的策略则无法进行泛化。最后，文章证明了SAR在更广泛的高维控制问题上的一致性，使用了机器人搅拌任务集和全身人形机器人步行任务。Abstract
Learning effective continuous control policies in high-dimensional systems, including musculoskeletal agents, remains a significant challenge. Over the course of biological evolution, organisms have developed robust mechanisms for overcoming this complexity to learn highly sophisticated strategies for motor control. What accounts for this robust behavioral flexibility? Modular control via muscle synergies, i.e. coordinated muscle co-contractions, is considered to be one putative mechanism that enables organisms to learn muscle control in a simplified and generalizable action space. Drawing inspiration from this evolved motor control strategy, we use physiologically accurate human hand and leg models as a testbed for determining the extent to which a Synergistic Action Representation (SAR) acquired from simpler tasks facilitates learning more complex tasks. We find in both cases that SAR-exploiting policies significantly outperform end-to-end reinforcement learning. Policies trained with SAR were able to achieve robust locomotion on a wide set of terrains with high sample efficiency, while baseline approaches failed to learn meaningful behaviors. Additionally, policies trained with SAR on a multiobject manipulation task significantly outperformed (>70% success) baseline approaches (<20% success). Both of these SAR-exploiting policies were also found to generalize zero-shot to out-of-domain environmental conditions, while policies that did not adopt SAR failed to generalize. Finally, we establish the generality of SAR on broader high-dimensional control problems using a robotic manipulation task set and a full-body humanoid locomotion task. To the best of our knowledge, this investigation is the first of its kind to present an end-to-end pipeline for discovering synergies and using this representation to learn high-dimensional continuous control across a wide diversity of tasks.
摘要
学习高维系统中的连续控制策略是一项挑战。生物演化过程中，生物体发展出了一些强大的机制来解决这种复杂性，以学习高度复杂的动作控制策略。这种行为的灵活性来源于哪里？模块化控制通过肌肉同步，即肌肉共同收缩，是一种被认为是生物体学习肌肉控制的潜在机制。我们以人工手和脚模型作为测试平台，通过使用生理学准确的人工手和脚模型，确定使用SAR（Synergistic Action Representation）从更简单的任务中获得的策略是否可以帮助学习更复杂的任务。我们发现，在两个情况下，使用SAR策略的学习效果都高于端到端学习。SAR策略可以在各种不同的地形上实现稳定的移动，而基线方法无法学习有意义的行为。此外，在多个目标拼接任务中，使用SAR策略的成功率高于70%，而基线方法的成功率低于20%。同时，这两种SAR策略还能够适应不同的环境条件，而不使用SAR的策略无法适应。最后，我们通过使用机器人 manipulate 任务集和全身人形步行任务，证明SAR在更广泛的高维控制问题上具有普遍性。据我们所知，这是首次对高维连续控制问题的整体解决方案。
INT-FP-QSim: Mixed Precision and Formats For Large Language Models and Vision Transformers
results: 这篇论文通过使用 INT-FP-QSim 模拟器，评估了不同数值精度和格式下 LLM 和 VT 的性能，并比较了最近提出的 Adaptive Block Floating Point、SmoothQuant、GPTQ 和 RPTQ 等方法的影响。Abstract
The recent rise of large language models (LLMs) has resulted in increased efforts towards running LLMs at reduced precision. Running LLMs at lower precision supports resource constraints and furthers their democratization, enabling users to run billion-parameter LLMs on their personal devices. To supplement this ongoing effort, we propose INT-FP-QSim: an open-source simulator that enables flexible evaluation of LLMs and vision transformers at various numerical precisions and formats. INT-FP-QSim leverages existing open-source repositories such as TensorRT, QPytorch and AIMET for a combined simulator that supports various floating point and integer formats. With the help of our simulator, we survey the impact of different numerical formats on the performance of LLMs and vision transformers at 4-bit weights and 4-bit or 8-bit activations. We also compare recently proposed methods like Adaptive Block Floating Point, SmoothQuant, GPTQ and RPTQ on the model performances. We hope INT-FP-QSim will enable researchers to flexibly simulate models at various precisions to support further research in quantization of LLMs and vision transformers.
摘要
最近的大语言模型（LLMs）的崛起导致了减少精度下运行的努力的增加。在减少精度下运行 LLMs 支持资源限制和普及，使用者可以在个人设备上运行 billion-parameter LLMs。为此，我们提议 INT-FP-QSim：一个开源的 simulator，可以在不同的数字精度和格式下灵活评估 LLMs 和视transformers。INT-FP-QSim 利用现有的开源库 such as TensorRT, QPytorch 和 AIMET，构建了一个集成的 simulator，支持多种浮点数和整数格式。我们使用 INT-FP-QSim，对 LLMs 和视transformers 的不同数字格式的影响进行了评估，并对最近提出的方法 like Adaptive Block Floating Point, SmoothQuant, GPTQ 和 RPTQ 进行了比较。我们希望 INT-FP-QSim 能够帮助研究人员在不同的精度下灵活模拟模型，以支持更多的 LLMS 和视transformers 的量化研究。
Fermat Distances: Metric Approximation, Spectral Convergence, and Clustering Algorithms
results: 这个论文证明了费马距离在小邻域内收敛到其连续类比中，收敛速率取决于数据的内在维度和权重参数。此外，这个论文还证明了基于费马距离的图 Laplacian 在维度感知的数据上的收敛性。Abstract
We analyze the convergence properties of Fermat distances, a family of density-driven metrics defined on Riemannian manifolds with an associated probability measure. Fermat distances may be defined either on discrete samples from the underlying measure, in which case they are random, or in the continuum setting, in which they are induced by geodesics under a density-distorted Riemannian metric. We prove that discrete, sample-based Fermat distances converge to their continuum analogues in small neighborhoods with a precise rate that depends on the intrinsic dimensionality of the data and the parameter governing the extent of density weighting in Fermat distances. This is done by leveraging novel geometric and statistical arguments in percolation theory that allow for non-uniform densities and curved domains. Our results are then used to prove that discrete graph Laplacians based on discrete, sample-driven Fermat distances converge to corresponding continuum operators. In particular, we show the discrete eigenvalues and eigenvectors converge to their continuum analogues at a dimension-dependent rate, which allows us to interpret the efficacy of discrete spectral clustering using Fermat distances in terms of the resulting continuum limit. The perspective afforded by our discrete-to-continuum Fermat distance analysis leads to new clustering algorithms for data and related insights into efficient computations associated to density-driven spectral clustering. Our theoretical analysis is supported with numerical simulations and experiments on synthetic and real image data.
摘要
我们分析 Ferma 距离的归一化性质，这是在里曼纹理上定义的一家density-driven métrique中的一个家族。Ferma 距离可以在粒子样本上定义，在这种情况下它们是随机的，也可以在 kontinuum 设置下定义，在这种情况下它们由 geodesics 下的 density-distorted Riemannian métrique 引起。我们证明了粒子样本上的 discrete Fermat 距离在小邻域内与其 kontinuum 同构中的analogues converge 到一个具体的rate，这个rate取决于数据的内在维度和 density weighting 参数的值。我们使用了新的几何和统计学理由，包括非均匀的density和拐弯的 Domian，来证明这一结论。我们的结果被用来证明基于 discrete Fermat 距离的图 Laplacian 的数据 clustering 算法是可靠的。具体来说，我们证明了 discrete eigenvalues 和 eigenvectors 在一定的维度上与其 kontinuum 同构中的analogues converge 到一个具体的rate，这使得我们可以根据continuum limit 来 интерпретирова discrete spectral clustering 的效果。我们的理论分析得到了 numérical simulations 和实验数据的支持，并提供了新的 clustering 算法和相关的办法。
Equivariant Single View Pose Prediction Via Induced and Restricted Representations
results: 这篇论文提出了一种新的算法，可以在三维世界中从二维图像中学习对象的姿态。该算法在三个不同的 pose 预测任务上进行了测试，并在PASCAL3D+和SYMSOL pose estimation任务上达到了最高的测试精度。Abstract
Learning about the three-dimensional world from two-dimensional images is a fundamental problem in computer vision. An ideal neural network architecture for such tasks would leverage the fact that objects can be rotated and translated in three dimensions to make predictions about novel images. However, imposing SO(3)-equivariance on two-dimensional inputs is difficult because the group of three-dimensional rotations does not have a natural action on the two-dimensional plane. Specifically, it is possible that an element of SO(3) will rotate an image out of plane. We show that an algorithm that learns a three-dimensional representation of the world from two dimensional images must satisfy certain geometric consistency properties which we formulate as SO(2)-equivariance constraints. We use the induced and restricted representations of SO(2) on SO(3) to construct and classify architectures which satisfy these geometric consistency constraints. We prove that any architecture which respects said consistency constraints can be realized as an instance of our construction. We show that three previously proposed neural architectures for 3D pose prediction are special cases of our construction. We propose a new algorithm that is a learnable generalization of previously considered methods. We test our architecture on three pose predictions task and achieve SOTA results on both the PASCAL3D+ and SYMSOL pose estimation tasks.
摘要
学习三维世界从二维图像中是计算机视觉的基本问题。理想的神经网络架构 для此类任务应该利用对象可以在三维空间中旋转和平移，以便对 novel 图像进行预测。然而，对于二维输入，强制 SO(3) 同态性是困难的，因为三维 rotate 操作没有自然的二维平面上的行为。我们表明，一个学习三维世界的算法从二维图像中获得三维表示的世界，必须满足某些几何一致性要求，我们将这些要求表述为 SO(2) 同态性约束。我们使用 SO(2) 在 SO(3) 上的启发和受限表示来设计和分类架构，并证明任何满足这些几何一致性要求的架构都可以通过我们的构造实现。我们表明，前面已经提出的三个神经网络架构 для 3D 姿态预测都是我们的构造的特例。我们提出了一种新的算法，它是learnable的，并且是已知的方法的普适化。我们对三个姿态预测任务进行测试，并在 PASCAL3D+ 和 SYMSOL 姿态预测任务上达到了最高的成绩。
Scalable Membership Inference Attacks via Quantile Regression
paper_authors: Martin Bertran, Shuai Tang, Michael Kearns, Jamie Morgenstern, Aaron Roth, Zhiwei Steven Wu
for: The paper is written for discussing a new class of membership inference attacks that are competitive with state-of-the-art shadow model attacks but require substantially less compute.
methods: The paper uses quantile regression on the distribution of confidence scores induced by the model under attack on points that are not used in training as the attack method.
results: The paper shows the efficacy of this approach in an extensive series of experiments on various datasets and model architectures, demonstrating that the proposed attack is competitive with state-of-the-art shadow model attacks while requiring less compute and being truly “black-box”.Abstract
Membership inference attacks are designed to determine, using black box access to trained models, whether a particular example was used in training or not. Membership inference can be formalized as a hypothesis testing problem. The most effective existing attacks estimate the distribution of some test statistic (usually the model's confidence on the true label) on points that were (and were not) used in training by training many \emph{shadow models} -- i.e. models of the same architecture as the model being attacked, trained on a random subsample of data. While effective, these attacks are extremely computationally expensive, especially when the model under attack is large. We introduce a new class of attacks based on performing quantile regression on the distribution of confidence scores induced by the model under attack on points that are not used in training. We show that our method is competitive with state-of-the-art shadow model attacks, while requiring substantially less compute because our attack requires training only a single model. Moreover, unlike shadow model attacks, our proposed attack does not require any knowledge of the architecture of the model under attack and is therefore truly ``black-box". We show the efficacy of this approach in an extensive series of experiments on various datasets and model architectures.
摘要
域名推测攻击是用于判断一个特定示例是否在训练中使用了黑盒访问已经训练过的模型。域名推测可以形式化为一个假设测试问题。现有最有效的攻击方法是通过训练多个“陌生模型”（即与被攻击模型相同的架构的模型，被训练于Random subsets of data）来估计模型在真实标签上的信任度分布。然而，这些攻击非常 computationally expensive，特别是当模型被攻击时很大。我们介绍了一种新的攻击方法，基于模型下发的信任分布中的分值回归。我们表明，我们的方法与现有的陌生模型攻击相比，需要更少的计算资源，因为我们的攻击只需要训练一个模型。此外，我们的提议的攻击方法不需要知道模型下发的架构，因此是真正的“黑盒”攻击。我们在多个数据集和模型架构上进行了广泛的实验，证明了这种方法的有效性。