for: The paper is written to tackle the challenges of generalization, long-term memory, and meta-learning in in-context Reinforcement Learning (RL) agents.
methods: The paper proposes a new in-context RL agent called AMAGO, which uses sequence models and off-policy learning to overcome the limitations of previous approaches.
results: The paper demonstrates the strong performance of AMAGO in meta-RL and long-term memory domains, and shows that it can solve goal-conditioned problems with challenging exploration. Additionally, the paper introduces a novel hindsight relabeling scheme that allows AMAGO to solve open-world domains.Abstract
We introduce AMAGO, an in-context Reinforcement Learning (RL) agent that uses sequence models to tackle the challenges of generalization, long-term memory, and meta-learning. Recent works have shown that off-policy learning can make in-context RL with recurrent policies viable. Nonetheless, these approaches require extensive tuning and limit scalability by creating key bottlenecks in agents' memory capacity, planning horizon, and model size. AMAGO revisits and redesigns the off-policy in-context approach to successfully train long-sequence Transformers over entire rollouts in parallel with end-to-end RL. Our agent is uniquely scalable and applicable to a wide range of problems. We demonstrate its strong performance empirically in meta-RL and long-term memory domains. AMAGO's focus on sparse rewards and off-policy data also allows in-context learning to extend to goal-conditioned problems with challenging exploration. When combined with a novel hindsight relabeling scheme, AMAGO can solve a previously difficult category of open-world domains, where agents complete many possible instructions in procedurally generated environments. We evaluate our agent on three goal-conditioned domains and study how its individual improvements connect to create a generalist policy.
摘要
我们介绍AMAGO,一个内Context Reinforcement Learning(RL)代理人,使用序列模型解决通用化、长期记忆和元学习的挑战。现有研究表明,离政RL可以使内Context RL with recurrent policies成为可能。然而,这些方法需要广泛的调整和限制数据容量、观察 horizon和模型大小,导致代理人的可扩展性和应用范围受限。AMAGO重新评估和重新设计了离政内Context Approach,成功地在整个推套中平行训练长序Transformer,并且具有广泛适用性。我们在Meta-RL和长期记忆领域 empirically 显示了它的强大表现。AMAGO的专注点在于罕见的 reward和离政数据也使内Context learning扩展到目标条件下的问题。当与一个新的预测重新标示方案相结合时,AMAGO可以解决一些过去Difficult的开放世界领域,其中代理人完成了许多可能的指令在生成的环境中。我们在三个目标条件下评估了我们的代理人,并研究了它们个别的改进如何相互连接,创建一个通用的政策。
Theoretical Evaluation of Asymmetric Shapley Values for Root-Cause Analysis
results: 研究发现,在某些情况下,ASV可能会产生Counter-intuitive的解释结果,这可能会导致模型预测中的根本原因分析错误。此外,研究还发现了一些特定的模型家族,如泛型加itive模型(GAM),在这些家族中,ASV具有愉悦的性质。Abstract
In this work, we examine Asymmetric Shapley Values (ASV), a variant of the popular SHAP additive local explanation method. ASV proposes a way to improve model explanations incorporating known causal relations between variables, and is also considered as a way to test for unfair discrimination in model predictions. Unexplored in previous literature, relaxing symmetry in Shapley values can have counter-intuitive consequences for model explanation. To better understand the method, we first show how local contributions correspond to global contributions of variance reduction. Using variance, we demonstrate multiple cases where ASV yields counter-intuitive attributions, arguably producing incorrect results for root-cause analysis. Second, we identify generalized additive models (GAM) as a restricted class for which ASV exhibits desirable properties. We support our arguments by proving multiple theoretical results about the method. Finally, we demonstrate the use of asymmetric attributions on multiple real-world datasets, comparing the results with and without restricted model families using gradient boosting and deep learning models.
摘要
在这项工作中,我们研究非对称的雪平值(ASV),这是SHAP添加式本地解释方法的一种变体。ASV提出了 incorporating known causal relations between variables的方法,并且被视为测试模型预测中的不公正折衔测试。在前期文献中没有被研究过,放弃雪平值的对称性可能会导致模型解释中的counter-intuitive consequence。为了更好地理解这种方法,我们首先示出了local contributions与global contributions of variance reduction的对应关系。使用方差,我们展示了多种情况下,ASV可能会生成错误的root-cause分析结果。其次,我们认为Generalized Additive Models(GAM)是一种受限的模型家族,ASV在这种模型家族中具有恰当的性质。我们支持我们的 Argument by proving multiple theoretical results about the method。最后,我们在多个实际 dataset上使用非对称的贡献值,并与和 без Restricted model families using gradient boosting和deep learning模型进行比较。
Deep Reinforcement Learning with Explicit Context Representation
paper_authors: Francisco Munguia-Galeano, Ah-Hwee Tan, Ze Ji
for: solves complex computational problems with contextual information
methods: uses Iota explicit context representation (IECR) framework with contextual key frames (CKFs) and two loss functions
results: significantly outperforms state-of-the-art equivalents in five discrete environments with contextual informationAbstract
Reinforcement learning (RL) has shown an outstanding capability for solving complex computational problems. However, most RL algorithms lack an explicit method that would allow learning from contextual information. Humans use context to identify patterns and relations among elements in the environment, along with how to avoid making wrong actions. On the other hand, what may seem like an obviously wrong decision from a human perspective could take hundreds of steps for an RL agent to learn to avoid. This paper proposes a framework for discrete environments called Iota explicit context representation (IECR). The framework involves representing each state using contextual key frames (CKFs), which can then be used to extract a function that represents the affordances of the state; in addition, two loss functions are introduced with respect to the affordances of the state. The novelty of the IECR framework lies in its capacity to extract contextual information from the environment and learn from the CKFs' representation. We validate the framework by developing four new algorithms that learn using context: Iota deep Q-network (IDQN), Iota double deep Q-network (IDDQN), Iota dueling deep Q-network (IDuDQN), and Iota dueling double deep Q-network (IDDDQN). Furthermore, we evaluate the framework and the new algorithms in five discrete environments. We show that all the algorithms, which use contextual information, converge in around 40,000 training steps of the neural networks, significantly outperforming their state-of-the-art equivalents.
摘要
强化学习(RL)已经表现出解决复杂计算问题的惊人能力。然而,大多数RL算法缺乏显式的方法来学习上下文信息。人类通过上下文来识别环境中元素之间的征交和相互关系,以及如何避免 incorrect 行为。相反,RL Agent可能需要多达百步才能学习避免错误的决策。这篇论文提出了一个名为IECR(Iota Explicit Context Representation)的框架,该框架可以在精确的环境中提取上下文信息,并学习CKFs(上下文关键帧)的表示。此外,本文还提出了两个相对于上下文的产品函数损失函数。IECR框架的创新之处在于可以从环境中提取上下文信息,并学习CKFs的表示。我们验证了IECR框架,并开发了四种使用上下文学习的算法:Iota Deep Q-Network(IDQN)、Iota Double Deep Q-Network(IDDQN)、Iota Duelling Deep Q-Network(IDuDQN)和Iota Duelling Double Deep Q-Network(IDDDQN)。此外,我们还在五个精确环境中评估了IECR框架和这些算法。我们发现,所有使用上下文信息的算法在40,000步训练步骤后,可以快速并高效地学习,与当前最佳算法相比显著性能更高。
BONES: Near-Optimal Neural-Enhanced Video Streaming
results: 比对当前状态艺技术,BONES算法可以提高用户视频流程体验质量4%到13%,显示其在提高视频流程体验质量方面具有潜在的应用前景。Abstract
Accessing high-quality video content can be challenging due to insufficient and unstable network bandwidth. Recent advances in neural enhancement have shown promising results in improving the quality of degraded videos through deep learning. Neural-Enhanced Streaming (NES) incorporates this new approach into video streaming, allowing users to download low-quality video segments and then enhance them to obtain high-quality content without violating the playback of the video stream. We introduce BONES, an NES control algorithm that jointly manages the network and computational resources to maximize the quality of experience (QoE) of the user. BONES formulates NES as a Lyapunov optimization problem and solves it in an online manner with near-optimal performance, making it the first NES algorithm to provide a theoretical performance guarantee. Our comprehensive experimental results indicate that BONES increases QoE by 4% to 13% over state-of-the-art algorithms, demonstrating its potential to enhance the video streaming experience for users. Our code and data will be released to the public.
摘要
Accessing high-quality video content can be challenging due to insufficient and unstable network bandwidth. Recent advances in neural enhancement have shown promising results in improving the quality of degraded videos through deep learning. Neural-Enhanced Streaming (NES) incorporates this new approach into video streaming, allowing users to download low-quality video segments and then enhance them to obtain high-quality content without violating the playback of the video stream. We introduce BONES, an NES control algorithm that jointly manages the network and computational resources to maximize the quality of experience (QoE) of the user. BONES formulates NES as a Lyapunov optimization problem and solves it in an online manner with near-optimal performance, making it the first NES algorithm to provide a theoretical performance guarantee. Our comprehensive experimental results indicate that BONES increases QoE by 4% to 13% over state-of-the-art algorithms, demonstrating its potential to enhance the video streaming experience for users. Our code and data will be released to the public.Here's the text in Traditional Chinese:Accessing high-quality video content can be challenging due to insufficient and unstable network bandwidth. Recent advances in neural enhancement have shown promising results in improving the quality of degraded videos through deep learning. Neural-Enhanced Streaming (NES) incorporates this new approach into video streaming, allowing users to download low-quality video segments and then enhance them to obtain high-quality content without violating the playback of the video stream. We introduce BONES, an NES control algorithm that jointly manages the network and computational resources to maximize the quality of experience (QoE) of the user. BONES formulates NES as a Lyapunov optimization problem and solves it in an online manner with near-optimal performance, making it the first NES algorithm to provide a theoretical performance guarantee. Our comprehensive experimental results indicate that BONES increases QoE by 4% to 13% over state-of-the-art algorithms, demonstrating its potential to enhance the video streaming experience for users. Our code and data will be released to the public.
Evaluation of feature selection performance for identification of best effective technical indicators on stock market price prediction
results: 研究发现,每种 wrapper 特征选择方法都有不同的结果,与不同的机器学习方法相关。ridge 和 LR 估计器,单独使用和与 wrapper 特征选择方法结合使用,在所有评价标准下得到了最佳股票市场预测结果。Abstract
Due to the influence of many factors, including technical indicators on stock market prediction, feature selection is important to choose the best indicators. One of the feature selection methods that consider the performance of models during feature selection is the wrapper feature selection method. The aim of this research is to identify a combination of the best stock market indicators through feature selection to predict the stock market price with the least error. In order to evaluate the impact of wrapper feature selection techniques on stock market prediction, in this paper SFS and SBS with 10 estimators and 123 technical indicators have been examined on the last 13 years of Apple Company. Also, by the proposed method, the data created by the 3-day time window were converted to the appropriate input for regression methods. Based on the results observed: (1) Each wrapper feature selection method has different results with different machine learning methods, and each method is more correlated with a specific set of technical indicators of the stock market. (2) Ridge and LR estimates alone, and with two methods of the wrapper feature selection, namely SFS and SBS; They had the best results with all assessment criteria for market forecast. (3)The Ridge and LR method with all the R2, MSE, RMSE, MAE and MAPE have the best stock market prediction results. Also, the MLP Regression Method, along with the Sequential Forwards Selection and the MSE, had the best performance. SVR regression, along with the SFS and the MSE, has improved greatly compared to the SVR regression with all indicators. (4) It was also observed that different features are selected by different ML methods with different evaluation parameters. (5) Most ML methods have used the Squeeze_pro, Percentage Price Oscillator, Thermo, Decay, Archer On-Balance Volume, Bollinger Bands, Squeeze and Ichimoku indicator.
摘要
因为多种因素的影响,包括技术指标在股票市场预测中,特征选择是重要的。本研究的目标是通过特征选择选择最佳的股票市场指标,以预测股票市场价格的误差最小。为了评估包装特征选择技术对股票市场预测的影响,本文在Apple公司上评估了13年的数据。具体来说,通过提posed方法,将3天时窗内的数据转换为适合回归方法的输入。根据结果所见:1. 每种包装特征选择方法都有不同的结果,与不同的机器学习方法相关。每种方法更加相关于股票市场技术指标的特定集。2. Ridge和LR估计独立,以及使用SFS和SBS两种包装特征选择方法时,在所有评价标准中表现最佳。3. Ridge和LR方法与所有评价标准(R2、MSE、RMSE、MAE和MAPE)表现最佳。此外,MLP回归方法,结合Sequential Forwards Selection和MSE,也表现出色。4. Observation showed that different ML methods select different features with different evaluation parameters.5. Most ML methods use Squeeze_pro、Percentage Price Oscillator、Thermo、Decay、Archer On-Balance Volume、Bollinger Bands、Squeeze和Ichimoku指标。
Towards Deep Learning Models Resistant to Transfer-based Adversarial Attacks via Data-centric Robust Learning
results: DRL在黑盒Robustness方面比PGD-AT、TRADES、EAT和FAT等常用AT技术表现出色,并且可以与多种数据增强和损失规则结合使用,以提高防御性。Abstract
Transfer-based adversarial attacks raise a severe threat to real-world deep learning systems since they do not require access to target models. Adversarial training (AT), which is recognized as the strongest defense against white-box attacks, has also guaranteed high robustness to (black-box) transfer-based attacks. However, AT suffers from heavy computational overhead since it optimizes the adversarial examples during the whole training process. In this paper, we demonstrate that such heavy optimization is not necessary for AT against transfer-based attacks. Instead, a one-shot adversarial augmentation prior to training is sufficient, and we name this new defense paradigm Data-centric Robust Learning (DRL). Our experimental results show that DRL outperforms widely-used AT techniques (e.g., PGD-AT, TRADES, EAT, and FAT) in terms of black-box robustness and even surpasses the top-1 defense on RobustBench when combined with diverse data augmentations and loss regularizations. We also identify other benefits of DRL, for instance, the model generalization capability and robust fairness.
摘要
transfer-based adversarial attacks pose a severe threat to real-world deep learning systems, as they do not require access to target models. adversarial training (AT), which is recognized as the strongest defense against white-box attacks, has also guaranteed high robustness to (black-box) transfer-based attacks. however, AT suffers from heavy computational overhead, as it optimizes adversarial examples during the entire training process. in this paper, we demonstrate that such heavy optimization is not necessary for AT against transfer-based attacks. instead, a one-shot adversarial augmentation prior to training is sufficient, and we name this new defense paradigm data-centric robust learning (drl). our experimental results show that drl outperforms widely-used at techniques (e.g., pgd-at, trades, eat, and fat) in terms of black-box robustness and even surpasses the top-1 defense on robustbench when combined with diverse data augmentations and loss regularizations. we also identify other benefits of drl, such as model generalization capability and robust fairness.
Score-Based Methods for Discrete Optimization in Deep Learning
paper_authors: Eric Lei, Arman Adibi, Hamed Hassani
for: 这 paper 是用于解决深度学习任务中的离散优化问题的。
methods: 这 paper 使用了一种分数函数方法来解决这些问题,该方法使用了一个分数函数作为目标函数的代理,并使用了隐藏变量的嵌入和自动导函数框架来并行计算反向传播。
results: 该 paper 的实验表明,在对抗式集成分类任务中,该方法可以实现一个更好的平衡点,即快速并且解决高维数据的问题。Abstract
Discrete optimization problems often arise in deep learning tasks, despite the fact that neural networks typically operate on continuous data. One class of these problems involve objective functions which depend on neural networks, but optimization variables which are discrete. Although the discrete optimization literature provides efficient algorithms, they are still impractical in these settings due to the high cost of an objective function evaluation, which involves a neural network forward-pass. In particular, they require $O(n)$ complexity per iteration, but real data such as point clouds have values of $n$ in thousands or more. In this paper, we investigate a score-based approximation framework to solve such problems. This framework uses a score function as a proxy for the marginal gain of the objective, leveraging embeddings of the discrete variables and speed of auto-differentiation frameworks to compute backward-passes in parallel. We experimentally demonstrate, in adversarial set classification tasks, that our method achieves a superior trade-off in terms of speed and solution quality compared to heuristic methods.
摘要
几乎所有深度学习任务中都会遇到离散优化问题,即使神经网络通常处理连续数据。这类问题中的目标函数取决于神经网络,但优化变量是离散的。虽然离散优化文献中提供了高效的算法,但它们在这些设置中仍然不实用,因为目标函数评估的成本高,需要对神经网络进行前进传播,这需要 $O(n)$ 复杂度每次迭代。例如,实际数据如点云可能有 thousands 或更多的值。在这篇论文中,我们研究了一种分数函数近似框架,用于解决这些问题。这个框架使用分数函数作为目标函数的代理,利用离散变量的嵌入和自动导数框架来并行计算反向传播。我们在随机设置中的对抗性分类任务中实验ally示出,我们的方法可以在速度和解决质量之间取得优化的负号比例。
Empower Text-Attributed Graphs Learning with Large Language Models (LLMs)
methods: 使用 Large Language Models (LLMs) 提取标签中的Semantic信息,并生成相应的类别 exemplars,然后使用边预测器捕捉原始数据中的结构信息,并将新生成的样本纳入原始图像中。
results: 对ogbn-arxiv dataset进行了广泛的实验,并显示了在1-shot情况下对基eline模型的76%提升。Abstract
Text-attributed graphs have recently garnered significant attention due to their wide range of applications in web domains. Existing methodologies employ word embedding models for acquiring text representations as node features, which are subsequently fed into Graph Neural Networks (GNNs) for training. Recently, the advent of Large Language Models (LLMs) has introduced their powerful capabilities in information retrieval and text generation, which can greatly enhance the text attributes of graph data. Furthermore, the acquisition and labeling of extensive datasets are both costly and time-consuming endeavors. Consequently, few-shot learning has emerged as a crucial problem in the context of graph learning tasks. In order to tackle this challenge, we propose a lightweight paradigm called ENG, which adopts a plug-and-play approach to empower text-attributed graphs through node generation using LLMs. Specifically, we utilize LLMs to extract semantic information from the labels and generate samples that belong to these categories as exemplars. Subsequently, we employ an edge predictor to capture the structural information inherent in the raw dataset and integrate the newly generated samples into the original graph. This approach harnesses LLMs for enhancing class-level information and seamlessly introduces labeled nodes and edges without modifying the raw dataset, thereby facilitating the node classification task in few-shot scenarios. Extensive experiments demonstrate the outstanding performance of our proposed paradigm, particularly in low-shot scenarios. For instance, in the 1-shot setting of the ogbn-arxiv dataset, ENG achieves a 76% improvement over the baseline model.
摘要
文本拥有Graph neural networks (GNNs) 在网络领域中得到了广泛的应用,现有的方法使用word embedding模型来获取文本表示,然后将其传递给GNNs进行训练。在大型自然语言模型(LLMs)的出现之后,这些 модели的强大能力在信息检索和文本生成中得到了应用,可以大幅提高文本特征。然而,收集和标注大量数据都是成本和时间consuming的任务。因此,几拍学习成为了图学习任务中的一个关键问题。为解决这个问题,我们提出了一种轻量级的方法called ENG,它采用了一种插件式的方法来授权文本拥有Graph中的节点生成。具体来说,我们使用LLMs来提取标签中的semantic信息,并生成符合这些类别的样本作为示例。然后,我们使用边预测器来捕捉原始数据中的结构信息,并将新生成的样本与原始图 Integrate into the graph。这种方法利用了LLMs来增强类别信息,无需修改原始数据,因此可以轻松地在几拍学习场景下进行节点分类。我们的实验表明,ENG方法在低shot场景下表现出色,例如在ogbn-arxivdataset中的1拍 Setting中,ENG方法与基eline模型相比提高了76%。
Alpha Elimination: Using Deep Reinforcement Learning to Reduce Fill-In during Sparse Matrix Decomposition
methods: 这个论文使用了一种基于强化学习的叠合矩阵重新排序方法,即alphaElimination。这个方法以单玩家游戏的形式表现出叠合矩阵重新排序问题,并使用Monte-Carlo tree search和神经网络来找到最佳移动。
results: 这个论文的结果显示,alphaElimination 可以与现有的热点排序法相比,实现更好的填充避免,并且对叠合矩阵的分解过程和解释过程都有很好的影响。Abstract
A large number of computational and scientific methods commonly require decomposing a sparse matrix into triangular factors as LU decomposition. A common problem faced during this decomposition is that even though the given matrix may be very sparse, the decomposition may lead to a denser triangular factors due to fill-in. A significant fill-in may lead to prohibitively larger computational costs and memory requirement during decomposition as well as during the solve phase. To this end, several heuristic sparse matrix reordering methods have been proposed to reduce fill-in before the decomposition. However, finding an optimal reordering algorithm that leads to minimal fill-in during such decomposition is known to be a NP-hard problem. A reinforcement learning based approach is proposed for this problem. The sparse matrix reordering problem is formulated as a single player game. More specifically, Monte-Carlo tree search in combination with neural network is used as a decision making algorithm to search for the best move in our game. The proposed method, alphaElimination is found to produce significantly lesser non-zeros in the LU decomposition as compared to existing state-of-the-art heuristic algorithms with little to no increase in overall running time of the algorithm. The code for the project will be publicly available here\footnote{\url{https://github.com/misterpawan/alphaEliminationPaper}.
摘要
许多计算和科学方法通常需要将稀疏矩阵分解为LU分解。在这个分解过程中,常见的问题是,即使给定矩阵很稀疏,但是分解可能会导致三角因子更加稠密,即fill-in问题。这种填充可能会导致计算成本和内存需求急剧增加。为解决这个问题,许多启发式稀疏矩阵重新排序方法已经被提出。然而,找到最优的重新排序算法,以使得在这个分解过程中避免填充,是一个NP困难问题。本文提出了一种基于强化学习的方法。将稀疏矩阵重新排序问题定义为单player游戏。具体来说,使用Monte-Carlo搜索和神经网络的决策算法来搜索最佳的移动。该方法被命名为alphaElimination,并发现它可以在LU分解中生成许多更少的非零元素,与现有的状态艺术算法相比,几乎没有增加总运行时间。代码将在以下链接公开:\footnote{\url{https://github.com/misterpawan/alphaEliminationPaper}。
Enhancing ML model accuracy for Digital VLSI circuits using diffusion models: A study on synthetic data generation
results: 我们的结果表明,扩散模型生成的人工数据和实际数据之间存在很close的相似性。我们验证了生成的数据质量,并证明了数据扩展确实有效地提高了VLSI设计中的预测性。Abstract
Generative AI has seen remarkable growth over the past few years, with diffusion models being state-of-the-art for image generation. This study investigates the use of diffusion models in generating artificial data generation for electronic circuits for enhancing the accuracy of subsequent machine learning models in tasks such as performance assessment, design, and testing when training data is usually known to be very limited. We utilize simulations in the HSPICE design environment with 22nm CMOS technology nodes to obtain representative real training data for our proposed diffusion model. Our results demonstrate the close resemblance of synthetic data using diffusion model to real data. We validate the quality of generated data, and demonstrate that data augmentation certainly effective in predictive analysis of VLSI design for digital circuits.
摘要
这篇研究探讨使用扩散模型来生成人工训练数据,以提高后续机器学习模型在电子遗传学 задача中的准确性。我们使用HSPICE设计环境,使用22nm CMOS技术架构,从实验中获得真实训练数据,以验证扩散模型的效果。我们的结果显示,扩散模型生成的 sintetic 数据和实际数据之间存在着类似的相似性。我们还证明了生成数据的质量,并证明了数据增强对适用于对数字遗传学设计的预测分析有效。Note:* "扩散模型" (diffusion model) refers to a type of generative model that generates data by iteratively refining a random noise vector until it matches the target data distribution.* "HSPICE" is a circuit simulator that is widely used in the field of electronic design automation.* "CMOS" (Complementary Metal-Oxide-Semiconductor) is a technology node that is commonly used in the fabrication of integrated circuits.* "适用" (suitable) means appropriate or applicable.* "预测分析" (predictive analysis) refers to the use of statistical or machine learning techniques to forecast the behavior of a system or process.
XRMDN: A Recurrent Mixture Density Networks-based Architecture for Short-Term Probabilistic Demand Forecasting in Mobility-on-Demand Systems with High Volatility
results: 根据实验结果,XRMDN比三种参考模型(包括统计、机器学习和深度学习模型)在三个评估指标上表现更好,特别是在具有强波动性的需求预测中。此外,XRMDN还可以帮助优化MoD系统中的其他应用问题,例如在不确定性下进行优化。Abstract
In real Mobility-on-Demand (MoD) systems, demand is subject to high and dynamic volatility, which is difficult to predict by conventional time-series forecasting approaches. Most existing forecasting approaches yield the point value as the prediction result, which ignores the uncertainty that exists in the forecasting result. This will lead to the forecasting result severely deviating from the true demand value due to the high volatility existing in demand. To fill the gap, we propose an extended recurrent mixture density network (XRMDN), which extends the weight and mean neural networks to recurrent neural networks. The recurrent neurons for mean and variance can capture the trend of the historical data-series data, which enables a better forecasting result in dynamic and high volatility. We conduct comprehensive experiments on one taxi trip record and one bike-sharing real MoD data set to validate the performance of XRMDN. Specifically, we compare our model to three types of benchmark models, including statistical, machine learning, and deep learning models on three evaluation metrics. The validation results show that XRMDN outperforms the three groups of benchmark models in terms of the evaluation metrics. Most importantly, XRMDN substantially improves the forecasting accuracy with the demands in strong volatility. Last but not least, this probabilistic demand forecasting model contributes not only to the demand prediction in MoD systems but also to other optimization application problems, especially optimization under uncertainty, in MoD applications.
摘要
真实的流动性-on-需求(MoD)系统中的需求受到高度和动态的不稳定性影响,这些影响难以预测通过传统时间序列预测方法。大多数现有预测方法只预测点值,忽略预测结果中存在的不确定性。这将导致预测结果与真实需求值严重不符,因为需求的高度不稳定。为了填补这个空白,我们提议一种扩展的循环混合密度网络(XRMDN)模型,扩展了权重和均值神经网络到循环神经网络。循环神经网络可以捕捉历史数据时系列数据的趋势,从而实现更好的预测结果在动态和高度不稳定的情况下。我们对一个出租车旅程记录和一个自行车分享真实MoD数据集进行了广泛的实验,以验证XRMDN的性能。具体来说,我们与三种参考模型进行比较,包括统计、机器学习和深度学习模型,并在三个评价指标上进行比较。验证结果显示,XRMDN在评价指标上都高于参考模型。此外,XRMDN在需求强度不稳定的情况下显著提高了预测精度。最后,这种probabilistic需求预测模型不仅有助于MoD系统中的需求预测,还有助于其他优化应用问题,尤其是在MoD应用中的不确定性优化问题。
Secure and Robust Communications for Cislunar Space Networks
results: 研究结果显示,使用机器学习算法的cislunar空间领域意识能力可以实现96%的准确性,并且显示出了这种方法的扎实性和可靠性。Abstract
There is no doubt that the Moon has become the center of interest for commercial and international actors. Over the past decade, the number of planned long-term missions has increased dramatically. This makes the establishment of cislunar space networks (CSNs) crucial to orchestrate uninterrupted communications between the Moon and Earth. However, there are numerous challenges, unknowns, and uncertainties associated with cislunar communications that may pose various risks to lunar missions. In this study, we aim to address these challenges for cislunar communications by proposing a machine learning-based cislunar space domain awareness (SDA) capability that enables robust and secure communications. To this end, we first propose a detailed channel model for selected cislunar scenarios. Secondly, we propose two types of interference that could model anomalies that occur in cislunar space and are so far known only to a limited extent. Finally, we discuss our cislunar SDA to work in conjunction with the spacecraft communication system. Our proposed cislunar SDA, involving heuristic learning capabilities with machine learning algorithms, detects interference models with over 96% accuracy. The results demonstrate the promising performance of our cislunar SDA approach for secure and robust cislunar communication.
摘要
<>将文本翻译成简化中文。<>月球已成为商业和国际行动者的中心,过去一代,计划的长期任务数量有所增加。这使得在月球和地球之间建立cislunar空间网络(CSN)变得非常重要,以确保无间断的通信。然而,cislunar通信存在许多挑战、未知和不确定性,这些风险可能对月球任务产生影响。在这种情况下,我们提出了一种基于机器学习的cislunar空间领域意识(SDA)能力,以确保安全和可靠的通信。为此,我们首先提出了选择的cislunar场景下的通道模型。其次,我们提出了两种可能出现在cislunar空间中的干扰,这些干扰至今只有有限的知识。最后,我们讨论了我们的cislunar SDA如何与空间器通信系统结合使用。我们的提议的cislunar SDA,结合机器学习算法的启发学能力,可以检测干扰模型的准确率高达96%。结果表明我们的cislunar SDA方法在确保安全和可靠的cislunar通信方面表现出色。
MAGIC: Detecting Advanced Persistent Threats via Masked Graph Representation Learning
results: 本文在三个广泛使用的数据集上进行评估,结果显示MAGIC可以在所有测试场景中获得出色的探测结果,并与现有的APT探测方法相比,具有很大的性能优势。Abstract
Advance Persistent Threats (APTs), adopted by most delicate attackers, are becoming increasing common and pose great threat to various enterprises and institutions. Data provenance analysis on provenance graphs has emerged as a common approach in APT detection. However, previous works have exhibited several shortcomings: (1) requiring attack-containing data and a priori knowledge of APTs, (2) failing in extracting the rich contextual information buried within provenance graphs and (3) becoming impracticable due to their prohibitive computation overhead and memory consumption. In this paper, we introduce MAGIC, a novel and flexible self-supervised APT detection approach capable of performing multi-granularity detection under different level of supervision. MAGIC leverages masked graph representation learning to model benign system entities and behaviors, performing efficient deep feature extraction and structure abstraction on provenance graphs. By ferreting out anomalous system behaviors via outlier detection methods, MAGIC is able to perform both system entity level and batched log level APT detection. MAGIC is specially designed to handle concept drift with a model adaption mechanism and successfully applies to universal conditions and detection scenarios. We evaluate MAGIC on three widely-used datasets, including both real-world and simulated attacks. Evaluation results indicate that MAGIC achieves promising detection results in all scenarios and shows enormous advantage over state-of-the-art APT detection approaches in performance overhead.
摘要
高级攻击者所采用的持续攻击(APT)正在不断增长,对各种企业和机构构成极大的威胁。数据源推论分析在APT检测中得到了广泛应用。然而,先前的工作具有以下缺陷:(1)需要攻击数据和先验知识,(2)无法提取质量推论图中埋藏的详细信息,(3)因计算负担和内存占用过高而成为不可持续。在这篇论文中,我们介绍MAGIC,一种新的自动化和灵活的APT检测方法。MAGIC可以在不同的级别和水平进行多重粒度检测,并且可以通过匿名系统实体和行为模型来快速抽象出深度特征。通过检测异常系统行为,MAGIC可以同时进行系统实体层和批处理日志层APT检测。MAGIC特别地采用了掩码图表学习来模型无辜系统实体和行为,并通过异常检测方法来检测异常系统行为。MAGIC可以适应概念漂移,并在通用条件和检测场景下显示出优异表现。我们对三个广泛使用的数据集进行了评估,包括真实攻击和模拟攻击。评估结果表明,MAGIC在所有场景中具有扎实的检测效果,与当前APT检测方法相比,具有巨大的性能优势。
VFLAIR: A Research Library and Benchmark for Vertical Federated Learning
results: 本文对11种攻击和8种防御策略进行了实验性评估,并从不同的通信和模型分割设置中绘制了具体的发现和建议,以帮助实际应用中的VFL部署场景选择防御策略。Abstract
Vertical Federated Learning (VFL) has emerged as a collaborative training paradigm that allows participants with different features of the same group of users to accomplish cooperative training without exposing their raw data or model parameters. VFL has gained significant attention for its research potential and real-world applications in recent years, but still faces substantial challenges, such as in defending various kinds of data inference and backdoor attacks. Moreover, most of existing VFL projects are industry-facing and not easily used for keeping track of the current research progress. To address this need, we present an extensible and lightweight VFL framework VFLAIR (available at https://github.com/FLAIR-THU/VFLAIR), which supports VFL training with a variety of models, datasets and protocols, along with standardized modules for comprehensive evaluations of attacks and defense strategies. We also benchmark 11 attacks and 8 defenses performance under different communication and model partition settings and draw concrete insights and recommendations on the choice of defense strategies for different practical VFL deployment scenario.
摘要
Communication Compression for Byzantine Robust Learning: New Efficient Algorithms and Improved Rates
results: 作者在数学实验中测试了提议的方法,并证明了它们在非 convex 和 Polyak-Lojasiewicz 平坦函数中具有更高的收敛速率和更好的承受性。Abstract
Byzantine robustness is an essential feature of algorithms for certain distributed optimization problems, typically encountered in collaborative/federated learning. These problems are usually huge-scale, implying that communication compression is also imperative for their resolution. These factors have spurred recent algorithmic and theoretical developments in the literature of Byzantine-robust learning with compression. In this paper, we contribute to this research area in two main directions. First, we propose a new Byzantine-robust method with compression -- Byz-DASHA-PAGE -- and prove that the new method has better convergence rate (for non-convex and Polyak-Lojasiewicz smooth optimization problems), smaller neighborhood size in the heterogeneous case, and tolerates more Byzantine workers under over-parametrization than the previous method with SOTA theoretical convergence guarantees (Byz-VR-MARINA). Secondly, we develop the first Byzantine-robust method with communication compression and error feedback -- Byz-EF21 -- along with its bidirectional compression version -- Byz-EF21-BC -- and derive the convergence rates for these methods for non-convex and Polyak-Lojasiewicz smooth case. We test the proposed methods and illustrate our theoretical findings in the numerical experiments.
摘要
布日兹罗布特性是分布式优化问题中的重要特性,通常在协同学习/联邦学习中出现。这些问题通常很大规模,因此通信压缩也是必要的。这些因素在文献中促进了最近的算法和理论发展,包括Byzantine-robust学习与压缩。在这篇论文中,我们对这个研究领域进行了两个主要贡献。首先,我们提出了一种新的Byzantine-robust方法——Byz-DASHA-PAGE,并证明其在非对称和Polyak-Lojasiewicz细致优化问题中的更好的收敛率,小于前一个方法(Byz-VR-MARINA)的最佳理论收敛保证。其次,我们开发了第一种Byzantine-robust方法与通信压缩,并对其在非对称和Polyak-Lojasiewicz细致优化问题中的收敛率进行了分析。我们还开发了一种双向压缩版本——Byz-EF21-BC,并对其进行了数学分析。我们的实验证明了我们的理论发现。
FLrce: Efficient Federated Learning with Relationship-based Client Selection and Early-Stopping Strategy
results: FLrce提高了通信和计算效率,在更少的轮数下达到了相同的准确率,并且可以在提前终止FL来降低通信和计算资源的消耗。Abstract
Federated learning (FL) achieves great popularity in broad areas as a powerful interface to offer intelligent services to customers while maintaining data privacy. Nevertheless, FL faces communication and computation bottlenecks due to limited bandwidth and resource constraints of edge devices. To comprehensively address the bottlenecks, the technique of dropout is introduced, where resource-constrained edge devices are allowed to collaboratively train a subset of the global model parameters. However, dropout impedes the learning efficiency of FL under unbalanced local data distributions. As a result, FL requires more rounds to achieve appropriate accuracy, consuming more communication and computation resources. In this paper, we present FLrce, an efficient FL framework with a relationship-based client selection and early-stopping strategy. FLrce accelerates the FL process by selecting clients with more significant effects, enabling the global model to converge to a high accuracy in fewer rounds. FLrce also leverages an early stopping mechanism to terminate FL in advance to save communication and computation resources. Experiment results show that FLrce increases the communication and computation efficiency by 6% to 73.9% and 20% to 79.5%, respectively, while maintaining competitive accuracy.
摘要
federated learning (FL) 在各种领域得到了广泛的推广,作为一种保持数据隐私的强大接口,提供智能服务给客户。然而,FL 面临有限带宽和资源限制的边缘设备的通信和计算瓶颈。为了全面解决这些瓶颈,dropout技术被引入,允许有限资源的边缘设备共同训练全球模型参数的一部分。然而,dropout会降低FL在不均匀本地数据分布时的学习效率。因此,FL需要更多的轮次来达到适当的准确率,消耗更多的通信和计算资源。在这篇论文中,我们提出了FLrce,一个高效的FL框架,具有关系基于的客户选择和早期终止策略。FLrce 加速了FL 过程,选择有更大影响的客户,使全球模型更快地 converges 到高准确率。FLrce 还利用了早期终止机制,提前终止 FL,以保存通信和计算资源。实验结果表明,FLrce 可以提高通信和计算效率,在 6% 到 73.9% 和 20% 到 79.5% 之间,而且保持竞争性的准确率。
Dynamic Link Prediction for New Nodes in Temporal Graph Networks
results: 在三个公开的数据集上进行了实验,结果显示了这个模型的表现比以前的方法更好。具体来说,这个模型可以更好地预测新节点之间的连接,并且可以在几乎无预警情况下进行预测。Abstract
Modelling temporal networks for dynamic link prediction of new nodes has many real-world applications, such as providing relevant item recommendations to new customers in recommender systems and suggesting appropriate posts to new users on social platforms. Unlike old nodes, new nodes have few historical links, which poses a challenge for the dynamic link prediction task. Most existing dynamic models treat all nodes equally and are not specialized for new nodes, resulting in suboptimal performances. In this paper, we consider dynamic link prediction of new nodes as a few-shot problem and propose a novel model based on the meta-learning principle to effectively mitigate this problem. Specifically, we develop a temporal encoder with a node-level span memory to obtain a new node embedding, and then we use a predictor to determine whether the new node generates a link. To overcome the few-shot challenge, we incorporate the encoder-predictor into the meta-learning paradigm, which can learn two types of implicit information during the formation of the temporal network through span adaptation and node adaptation. The acquired implicit information can serve as model initialisation and facilitate rapid adaptation to new nodes through a fine-tuning process on just a few links. Experiments on three publicly available datasets demonstrate the superior performance of our model compared to existing state-of-the-art methods.
摘要
模拟 temporal networks 为新节点动态链接预测有多个实际应用,如为新用户提供相关的ITE推荐和社交平台上新用户的适当帖子推荐。与老节点不同,新节点具有少量历史链接,这对动态链接预测任务带来挑战。大多数现有的动态模型往往对所有节点进行等效处理,从而导致下OPTIMAL表现。在本文中,我们将动态链接预测新节点视为几枚shot问题,并提出一种基于元学习原则的新模型。具体来说,我们开发了一个包含节点级别span记忆的时间编码器,以获得新节点嵌入,然后使用一个预测器来判断新节点是否生成链接。为了解决几枚shot挑战,我们将编码器-预测器 integrate 到元学习 парадиг中,可以在形成 temporal network 过程中通过 span 适应和节点适应学习两种隐式信息。获得的隐式信息可以作为模型初始化,并且通过一些链接的精度适应来快速适应新节点。在三个公开的数据集上进行了实验,我们的模型比现有状态 искусственный方法表现出色。
results: 这paper的实验结果表明,使用 Pseudo-Bayesian Optimization 方法可以不仅确保优化过程的 convergence,还可以在高维 synthetic experiment、hyperparameter tuning 和机器人应用中实现更高的性能,并且可以与现有的state-of-the-art benchmarks相比,显示出更好的性能。Abstract
Bayesian Optimization is a popular approach for optimizing expensive black-box functions. Its key idea is to use a surrogate model to approximate the objective and, importantly, quantify the associated uncertainty that allows a sequential search of query points that balance exploitation-exploration. Gaussian process (GP) has been a primary candidate for the surrogate model, thanks to its Bayesian-principled uncertainty quantification power and modeling flexibility. However, its challenges have also spurred an array of alternatives whose convergence properties could be more opaque. Motivated by these, we study in this paper an axiomatic framework that elicits the minimal requirements to guarantee black-box optimization convergence that could apply beyond GP-related methods. Moreover, we leverage the design freedom in our framework, which we call Pseudo-Bayesian Optimization, to construct empirically superior algorithms. In particular, we show how using simple local regression, and a suitable "randomized prior" construction to quantify uncertainty, not only guarantees convergence but also consistently outperforms state-of-the-art benchmarks in examples ranging from high-dimensional synthetic experiments to realistic hyperparameter tuning and robotic applications.
摘要
bayesian 优化是一种广泛应用的优化方法,用于优化costly黑obox函数。其关键思想是使用一个surrogate模型来近似目标函数,同时能够量化相关的uncertainty,以实现sequential搜索。 Gaussian process(GP)因其 bayesian原理下的uncertainty量化能力和模型灵活性而成为主要候选人。然而,GP也存在一些挑战,这些挑战激发了一系列alternatives的发展。在这篇论文中,我们提出了一个axioms framework,该框架可以保证黑obox优化的收敛性,并且可以在GP相关方法之外应用。此外,我们利用了我们的框架的设计自由度,constructed一种empirically superior的算法。具体来说,我们使用了一个简单的local regression,并使用一种适当的“Randomized Prior”的构建来量化uncertainty。这不仅保证了收敛性,还可以在高维 synthetic experiments中consistently outperformstate-of-the-art benchmarks。
UniTime: A Language-Empowered Unified Model for Cross-Domain Time Series Forecasting
results: 实验结果显示UniTime模型能够提高现有模型的预测性能和零学习转移性能。Abstract
Multivariate time series forecasting plays a pivotal role in contemporary web technologies. In contrast to conventional methods that involve creating dedicated models for specific time series application domains, this research advocates for a unified model paradigm that transcends domain boundaries. However, learning an effective cross-domain model presents the following challenges. First, various domains exhibit disparities in data characteristics, e.g., the number of variables, posing hurdles for existing models that impose inflexible constraints on these factors. Second, the model may encounter difficulties in distinguishing data from various domains, leading to suboptimal performance in our assessments. Third, the diverse convergence rates of time series domains can also result in compromised empirical performance. To address these issues, we propose UniTime for effective cross-domain time series learning. Concretely, UniTime can flexibly adapt to data with varying characteristics. It also uses domain instructions and a Language-TS Transformer to offer identification information and align two modalities. In addition, UniTime employs masking to alleviate domain convergence speed imbalance issues. Our extensive experiments demonstrate the effectiveness of UniTime in advancing state-of-the-art forecasting performance and zero-shot transferability.
摘要
多变量时间序列预测在当代网络技术中扮演着关键角色。与传统方法不同,这项研究提出了跨领域模型的统一模型架构,跨越领域边界。然而,学习有效的跨领域模型存在以下挑战:首先,不同领域的数据特征存在差异,例如变量数量,这会限制现有模型的灵活性。其次,模型可能很难分辨不同领域的数据,导致预测性能下降。最后,时间序列领域的多样化速度也可能导致实际性能下降。为解决这些问题,我们提出了UniTime,一种可靠地适应数据特征变化的模型。具体来说,UniTime可以适应数据中的变量数量变化,并使用领域指令和语言-TS transformer来提供标识信息和对两种模式进行对应。此外,UniTime还使用屏蔽来缓解领域融合速度不平衡问题。我们的广泛实验表明UniTime可以提高预测性能和零代负荷传递性。
results: 根据实验结果,这种ensemble策略可以对于使用GAN-based differential privacy mechanisms(即生成机制)训练的下游模型,提高其在真实数据上的表现,包括精度和模型调整的方面。但是,这种策略不会对于使用margin-based或workload-based differential privacy mechanisms(即统计机制)训练的下游模型提高表现。Abstract
When machine learning models are trained on synthetic data and then deployed on real data, there is often a performance drop due to the distribution shift between synthetic and real data. In this paper, we introduce a new ensemble strategy for training downstream models, with the goal of enhancing their performance when used on real data. We generate multiple synthetic datasets by applying a differential privacy (DP) mechanism several times in parallel and then ensemble the downstream models trained on these datasets. While each synthetic dataset might deviate more from the real data distribution, they collectively increase sample diversity. This may enhance the robustness of downstream models against distribution shifts. Our extensive experiments reveal that while ensembling does not enhance downstream performance (compared with training a single model) for models trained on synthetic data generated by marginal-based or workload-based DP mechanisms, our proposed ensemble strategy does improve the performance for models trained using GAN-based DP mechanisms in terms of both accuracy and calibration of downstream models.
摘要
当机器学习模型在生成的数据上训练并在实际数据上部署时,通常会出现性能下降,这是因为生成的数据和实际数据的分布shift。在这篇论文中,我们介绍了一种新的集成策略,用于训练下游模型,以提高它们在实际数据上的表现。我们通过应用多次 diferencial privacy(DP)机制来生成多个生成的数据集,然后将这些数据集上训练的下游模型 ensemble。虽然每个生成的数据集可能更 deviation from the real data distribution,但它们的总体样本多样性可以提高下游模型对分布shift的Robustness。我们的广泛实验表明,对于基于marginal-based或workload-based DP机制生成的数据集,集成不会提高下游模型的性能(与单个模型训练相比),但是我们的提议的集成策略对基于GAN-based DP机制生成的数据集进行训练,可以提高模型的准确性和下游模型的Calibration。
SVM based Multiclass Classifier for Gait phase Classification using Shank IMU Sensor
results: 该方法可以高度准确地分类不同的步态阶段,准确率约为90.3%。Here’s a breakdown of each point:
for: 本研究旨在开发一种基于SVM多类分类的步态分类方法,以高精度地标识步态阶段,包括七个子阶段。 (The study aims to develop a gait phase classification method based on SVM multi-class classification, to accurately identify the gait phases, including seven sub-phases.)
methods: 该方法使用个体IMU传感器数据,如膝盖加速度x、y、z和膝盖陀螺x,作为特征进行分类。 (The method uses individual IMU sensor data, such as shank acceleration x, y, and z, and knee angles, as features for classification.)
results: 该方法可以高度准确地分类不同的步态阶段,准确率约为90.3%。 (The method can accurately classify different gait phases with an accuracy of approximately 90.3%.)Abstract
In this study, a gait phase classification method based on SVM multiclass classification is introduced, with a focus on the precise identification of the stance and swing phases, which are further subdivided into seven phases. Data from individual IMU sensors, such as Shank Acceleration X, Y, Z, Shank Gyro X, and Knee Angles, are used as features in this classification model. The suggested technique successfully classifies the various gait phases with a significant accuracy of about 90.3%. Gait phase classification is crucial, especially in the domains of exoskeletons and prosthetics, where accurate identification of gait phases enables seamless integration with assistive equipment, improving mobility, stability, and energy economy. This study extends the study of gait and offers an effective method for correctly identifying gait phases from Shank IMU sensor data, with potential applications in biomechanical research, exoskeletons, rehabilitation, and prosthetics.
摘要
在本研究中,基于SVM多类分类的步态分类方法被引入,强调精准地识别步态的不同阶段,这些阶段进一步细分为七个阶段。研究使用个体IMU传感器数据,如膝盖加速度x、y、z、膝盖陀螺x等,作为分类模型的特征。提议的技术成功地分类不同的步态阶段,准确率达到了约90.3%。步态分类在许多领域都非常重要,如外围机械助手和假肢,准确识别步态阶段可以帮助融合助手设备,提高 mobilidad、稳定性和能源经济。本研究对步态的研究进一步推广,并提供了基于膝盖IMU传感器数据的有效的步态分类方法,可能在生物机械研究、外围机械助手、rehabilitation和假肢领域得到应用。
Provably Fast Convergence of Independent Natural Policy Gradient for Markov Potential Games
results: 这个研究证明了,在 synthetic potential game 和 congestion game 中,独立NPG方法可以在 $\mathcal{O}(1/\epsilon)$ 迭代中达到 $\epsilon$-纳什平衡,超过之前的最佳结果 $\mathcal{O}(1/\epsilon^2)$ 迭代。Abstract
This work studies an independent natural policy gradient (NPG) algorithm for the multi-agent reinforcement learning problem in Markov potential games. It is shown that, under mild technical assumptions and the introduction of the \textit{suboptimality gap}, the independent NPG method with an oracle providing exact policy evaluation asymptotically reaches an $\epsilon$-Nash Equilibrium (NE) within $\mathcal{O}(1/\epsilon)$ iterations. This improves upon the previous best result of $\mathcal{O}(1/\epsilon^2)$ iterations and is of the same order, $\mathcal{O}(1/\epsilon)$, that is achievable for the single-agent case. Empirical results for a synthetic potential game and a congestion game are presented to verify the theoretical bounds.
摘要
SGA: A Graph Augmentation Method for Signed Graph Neural Networks
for: This paper is written for analyzing complex patterns in real-world signed graphs, and addressing three key challenges in SGNN-based signed graph representation learning: sparsity, unbalanced triangles, and lack of supplementary information.
methods: The paper proposes a novel Signed Graph Augmentation framework (SGA) that includes three main components: (1) using an SGNN model to encode the signed graph and extract latent structural information for candidate augmentation structures, (2) evaluating and selecting the most beneficial candidate samples for modifying the original training set, and (3) a novel augmentation perspective that assigns varying training difficulty to training samples.
results: The paper demonstrates significant improvements in performance across multiple benchmarks using the proposed SGA method, outperforming baselines by up to 22.2% in AUC, 33.3% in F1-binary, 48.8% in F1-micro, and 36.3% in F1-macro on six real-world datasets.Abstract
Signed Graph Neural Networks (SGNNs) are vital for analyzing complex patterns in real-world signed graphs containing positive and negative links. However, three key challenges hinder current SGNN-based signed graph representation learning: sparsity in signed graphs leaves latent structures undiscovered, unbalanced triangles pose representation difficulties for SGNN models, and real-world signed graph datasets often lack supplementary information like node labels and features. These constraints limit the potential of SGNN-based representation learning. We address these issues with data augmentation techniques. Despite many graph data augmentation methods existing for unsigned graphs, none are tailored for signed graphs. Our paper introduces the novel Signed Graph Augmentation framework (SGA), comprising three main components. First, we employ the SGNN model to encode the signed graph, extracting latent structural information for candidate augmentation structures. Second, we evaluate these candidate samples (edges) and select the most beneficial ones for modifying the original training set. Third, we propose a novel augmentation perspective that assigns varying training difficulty to training samples, enabling the design of a new training strategy. Extensive experiments on six real-world datasets (Bitcoin-alpha, Bitcoin-otc, Epinions, Slashdot, Wiki-elec, and Wiki-RfA) demonstrate that SGA significantly improves performance across multiple benchmarks. Our method outperforms baselines by up to 22.2% in AUC for SGCN on Wiki-RfA, 33.3% in F1-binary, 48.8% in F1-micro, and 36.3% in F1-macro for GAT on Bitcoin-alpha in link sign prediction.
摘要
Signed Graph Neural Networks (SGNNs) 是对实际中带有正负链接的签名图进行分析复杂模式的关键工具。然而,现有的SGNN模型在签名图表示学习中存在三大挑战:签名图中的稀疏性使得潜在结构未被发现,不均衡的triangle对SGNN模型进行表示带来挑战,而且现实中的签名图数据往往缺乏节点标签和特征信息。这些限制使SGNN-基于表示学习的潜力受限。我们通过数据扩充技术来解决这些问题。虽然现有许多对 unsigned 图进行数据扩充的方法,但是这些方法并没有适应签名图。我们的论文提出了一种新的签名图扩充框架(SGA),包括以下三个主要组成部分:1. 我们使用 SGNN 模型来编码签名图,提取签名图中的潜在结构信息作为候选扩充结构。2. 我们评估这些候选样本(边),并选择对原始训练集进行最有利的修改。3. 我们提出了一种新的增强训练方法,对各种训练样本分配不同的训练难度,以便设计更好的训练策略。我们在六个真实世界数据集(Bitcoin-alpha、Bitcoin-otc、Epinions、Slashdot、Wiki-elec和Wiki-RfA)进行了广泛的实验,结果表明,SGA 可以在多个 bench 上显著提高性能。我们的方法在 Wiki-RfA 上的 AUC 上比基eline 提高了22.2%,在 Bitcoin-alpha 上的 F1-binary、F1-micro 和 F1-macro 上提高了33.3%、48.8% 和 36.3%。
When Collaborative Filtering is not Collaborative: Unfairness of PCA for Recommendations
results: 这paper发现了PCA的两个下面机制,它们会导致推荐系统中的不公平性。此外,paper还提出了一种改进PCA的算法,即Item-Weighted PCA,可以更好地处理不同类型的项目。在一些假设的矩阵上,paper证明了Item-Weighted PCA使用特定的质量可以最小化一个媒体化错误度量。在实际数据上,这paper发现Item-Weighted PCA不仅可以提高总体推荐质量,还可以提高流行和不流行的项目。Abstract
We study the fairness of dimensionality reduction methods for recommendations. We focus on the established method of principal component analysis (PCA), which identifies latent components and produces a low-rank approximation via the leading components while discarding the trailing components. Prior works have defined notions of "fair PCA"; however, these definitions do not answer the following question: what makes PCA unfair? We identify two underlying mechanisms of PCA that induce unfairness at the item level. The first negatively impacts less popular items, due to the fact that less popular items rely on trailing latent components to recover their values. The second negatively impacts the highly popular items, since the leading PCA components specialize in individual popular items instead of capturing similarities between items. To address these issues, we develop a polynomial-time algorithm, Item-Weighted PCA, a modification of PCA that uses item-specific weights in the objective. On a stylized class of matrices, we prove that Item-Weighted PCA using a specific set of weights minimizes a popularity-normalized error metric. Our evaluations on real-world datasets show that Item-Weighted PCA not only improves overall recommendation quality by up to $0.1$ item-level AUC-ROC but also improves on both popular and less popular items.
摘要
我们研究了维度减少方法的公平性,特别是已知的主 componenets分析(PCA)方法。PCA方法可以找到缺失的特征并生成一个低级别的approximation,通过主要的特征来抛弃追随的特征。先前的研究已经定义了“公平PCA”的概念,但这些定义并没有回答以下问题:PCA方法是如何不公平的?我们认为PCA方法存在两种下面机制,导致item级别的不公平ness。首先,less popular items会受到负面影响,因为这些items rely on追随的特征来恢复其价值。其次,highly popular items会受到负面影响,因为leading PCA components会专注于个体受欢迎的items而不是捕捉items之间的相似性。为了解决这些问题,我们开发了一个幂时间算法,Item-Weighted PCA,这是PCA方法的修改。在一个简化的矩阵类型上,我们证明了Item-Weighted PCA使用的项目特定的权重在目标函数中具有最小化一个受欢迎度normalized error metric的性能。我们对实际数据进行评估,显示Item-Weighted PCA不仅提高了总的推荐质量,最高达0.1个item-level AUC-ROC,同时也提高了受欢迎和less popular items的质量。
Enhancing Column Generation by Reinforcement Learning-Based Hyper-Heuristic for Vehicle Routing and Scheduling Problems
results: 在 Vehicle Routing Problem with Time Windows 和 Bus Driver Scheduling Problem 两个典型的 combinatorial optimization 问题中,可以提高解得质量,最高减少总成本达 27.9% 和 15.4%,在相同或更少的计算时间内减少计算时间。Abstract
Column generation (CG) is a vital method to solve large-scale problems by dynamically generating variables. It has extensive applications in common combinatorial optimization, such as vehicle routing and scheduling problems, where each iteration step requires solving an NP-hard constrained shortest path problem. Although some heuristic methods for acceleration already exist, they are not versatile enough to solve different problems. In this work, we propose a reinforcement learning-based hyper-heuristic framework, dubbed RLHH, to enhance the performance of CG. RLHH is a selection module embedded in CG to accelerate convergence and get better integer solutions. In each CG iteration, the RL agent selects a low-level heuristic to construct a reduced network only containing the edges with a greater chance of being part of the optimal solution. In addition, we specify RLHH to solve two typical combinatorial optimization problems: Vehicle Routing Problem with Time Windows (VRPTW) and Bus Driver Scheduling Problem (BDSP). The total cost can be reduced by up to 27.9\% in VRPTW and 15.4\% in BDSP compared to the best lower-level heuristic in our tested scenarios, within equivalent or even less computational time. The proposed RLHH is the first RL-based CG method that outperforms traditional approaches in terms of solution quality, which can promote the application of CG in combinatorial optimization.
摘要
column generation (CG) 是一种重要的方法,用于解决大规模问题,通过动态生成变量。它在各种常见的 combinatorial optimization 中有广泛的应用,如车辆 Routing 和调度问题,每个迭代步骤都需要解决一个 NP-hard 约束短路问题。虽然一些启发法已经存在,但它们并不够 versatile enough 解决不同的问题。在这项工作中,我们提出了一种基于强化学习的 hyper-heuristic 框架,称为 RLHH,以提高 CG 的性能。RLHH 是 CG 中的一个选择模块,用于加速迭代和获得更好的整数解。在每个 CG 迭代中,RL Agent 将选择一个低级别启发,用于构建一个只包含有更高可能性成为优解的最佳解的减少网络。此外,我们将 RLHH 应用于两种典型的 combinatorial optimization 问题:车辆 Routing 问题 with Time Windows (VRPTW) 和 Bus Driver Scheduling 问题 (BDSP)。我们在测试场景中发现,RLHH 可以将总成本降低到 27.9% 以下,相对于最佳下级别启发,并且在相同或更少的计算时间内达成。我们的提案的 RLHH 是首个通过强化学习来超越传统方法的 CG 方法,可以提高 CG 在 combinatorial optimization 中的应用。