检索结果-内蒙古大学图书馆

Renewable Energy and Power Quality Journal 2023年 21卷 565-569页

作者： Uremović, N. Lukač, N. Sukic, P. Štumberger, G. Faculty of Electrical Engineering and Computer Science University of Maribor Koroška cesta 46 Maribor2000 Slovenia

Automated decision tools, such as advanced energy management systems, are required to involve the electrical grid users in energy flexibility services. This paper focuses on the prediction models as a substantial part of decision strategy in advanced energy management systems and on advanced energy management systems as a tool that supports the active involvement of electrical grid users in energy flexibility services. Prediction models' desired properties are self-establishing and self-adaptation, which require new solutions in data selection, filtering, processing and model learning. Some of these properties are investigated within this paper. © 2023, European Association for the Development of Renewable Energy, Environment and Power Quality (EA4EPQ). All rights reserved.

关键词： Energy management systems

来源：评论

学校读者我要写书评

暂无评论

Learning Multiagent Options for Tabular Reinforcement Learning using Factor Graphs

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2023年第5期4卷 1141-1153页

作者： Chen, Jiayu Chen, Jingdi Lan, Tian Aggarwal, Vaneet Purdue University School of Industrial Engineering West LafayetteIN47907 United States George Washington University Department of Electrical and Computer Engineering WashingtonDC20052 United States Purdue University School of Industrial Engineering The School of Electrical and Computer Engineering West LafayetteIN47907 United States King Abdullah University of Science and Technology Department of Computer Science Thuwal23955 Saudi Arabia

Covering option discovery has been developed to improve the exploration of reinforcement learning in single-agent scenarios, where only sparse reward signals are available. It aims to connect the most distant states identified through the Fiedler vector of the state transition graph. However, the approach cannot be directly extended to multiagent scenarios, since the joint state space grows exponentially with the number of agents, thus prohibiting efficient option computation. Existing research adopting options in multiagent scenarios still relies on single-agent algorithms and fails to directly discover joint options that can improve the connectivity of the joint state space. In this article, we propose a new algorithm to directly compute multiagent options with collaborative exploratory behaviors while still enjoying the ease of decomposition. Our key idea is to approximate the joint state space as the Kronecker product of individual agents' state spaces, based on which we can directly estimate the Fiedler vector of the joint state space using the Laplacian spectrum of individual agents' transition graphs. This decomposition enables us to efficiently construct multiagent joint options by encouraging agents to connect the subgoal joint states, which are corresponding to the minimum or maximum of the estimated joint Fiedler vector. Evaluation on multiagent collaborative tasks shows that our algorithm can successfully identify multiagent options and significantly outperforms prior works using single-agent options or no options, in terms of both faster exploration and higher cumulative rewards. © 2020 IEEE.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Core-Intermediate-Peripheral Index: Factor Analysis of Neighborhood and Shortest Paths-Based Centrality Metrics 7th

Core-Intermediate-Peripheral Index: Factor Analysis of Neigh...

引用

7th Computational Methods in Systems and Software, CoMeSySo 2023

作者： Meghanathan, Natarajan Department of Electrical and Computer Engineering and Computer Science Jackson State University 1400 Lynch Street JacksonMS39217 United States

ISBN: (纸本)9783031535512

We perform factor analysis on the raw data of the four major neighborhood and shortest paths-based centrality metrics (Degree, Eigenvector, Betweenness and Closeness) and propose a novel quantitative measure called the CoreIntermediate-Peripheral (CIP) Index to capture the extent with which a node could play the role of a core node (nodes at the center of a network with larger values for any centrality metric) vis-a-vis a peripheral node (nodes that exist at the periphery of a network with lower values for any centrality metric). We conduct factor analysis (varimax-based rotation of the Eigenvectors) on the transpose matrix of the raw centrality metrics dataset, with the node ids as features, under the hypothesis that there are two factors (core and peripheral) that drive the values incurred by the nodes with respect to the centrality metrics. We test our approach on a diverse suite of 12 complex real-world networks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Factor analysis

来源：评论

学校读者我要写书评

暂无评论

TOWARDS THE FUNDAMENTAL LIMITS OF KNOWLEDGE TRANSFER OVER FINITE DOMAINS 12

TOWARDS THE FUNDAMENTAL LIMITS OF KNOWLEDGE TRANSFER OVER FI...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Zhao, Qingyue Zhu, Banghua Department of Computer Science and Technology Tsinghua University China Department of Electrical Engineering and Computer Sciences University of California Berkeley United States

We characterize the statistical efficiency of knowledge transfer through n samples from a teacher to a probabilistic student classifier with input space S over labels A. We show that privileged information at three progressive levels accelerates the transfer. At the first level, only samples with hard labels are known, via which the maximum likelihood estimator attains the minimax rate √|S||A|/n. The second level has the teacher probabilities of sampled labels available in addition, which turns out to boost the convergence rate lower bound to |S||A|/n. However, under this second data acquisition protocol, minimizing a naive adaptation of the cross-entropy loss results in an asymptotically biased student. We overcome this limitation and achieve the fundamental limit by using a novel empirical variant of the squared error logit loss. The third level further equips the student with the soft labels (complete logits) on A given every sampled input, thereby provably enables the student to enjoy a rate |S|/n free of |A|. We find any Kullback-Leibler divergence minimizer to be optimal in the last case. Numerical simulations distinguish the four learners and corroborate our theory. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

A 0.75 to 1.5 GHz Reconfigurable Diplexer With Independently Controllable Center Frequency and Insertion Phase

引用

IEEE Transactions on Circuits and Systems II: Express Briefs 2024年第4期71卷 1924-1928页

作者： Yi, Chengfei Chen, Xiong Chi, Pei-Ling Yang, Tao The School of Electronic Science and Engineering University of Electronic Science and Technology of China Sichuan Chengdu611731 China The Department of Electrical and Computer Engineering National Chiao Tung University Hsinchu300 Taiwan

A novel reconfigurable diplexer with independently controllable center frequency and insertion phase is proposed in this brief. It simply consists of six coupled resonators and two non-resonating-nodes (NRNs). These six resonators are arranged in two triplet topologies and construct two typical third-order filters, forming the lower and upper channels in the diplexer, respectively. The NRNs are loaded at the receiver (RX) and transmitter (TX) ports. With the utilization of NRNs, the insertion phases of the upper and lower channels in the diplexer can be independently controlled. Optimization technique is used to obtain the coupling matrix of the diplexer, in which the NRN is considered to achieve a prespecified insertion phase. To validate the proposed idea, a microstrip prototype is fabricated and measured. Experiment results show that the center frequencies of lower and upper channels can be tuned from 0.75 GHz to 0.98 GHz, and from 1.23 GHz to 1.5 GHz, respectively. The insertion phases of both channels can be independently tuned from 0 to 120 degrees at each frequency tuning state. © 2023 IEEE.

关键词： Resonators

来源：评论

学校读者我要写书评

暂无评论

Use of Neighborhood-Based Bridge Node Centrality Tuple for Preferential Vaccination of Nodes to Reduce the Number of Infected Nodes in a Complex Real-World Network 1

引用

7th Computational Methods in Systems and Software, CoMeSySo 2023

作者： Meghanathan, Natarajan Burden, Kapri Robinson, Miah Department of Electrical and Computer Engineering and Computer Science Jackson State University 1400 Lynch Street JacksonMS39217 United States

ISBN: (数字)9783031535529

ISBN: (纸本)9783031535512

We investigate the use of the Neighborhood-based Bridge Node Centrality (NBNC) tuple to choose nodes for preferential vaccination so that such vaccinated nodes could provide herd immunity and reduce the spreading rate of infections in a complex real-world network. The NBNC tuple ranks nodes on the basis of the extent they serve as bridge nodes in a network. A node is a bridge node, if when removed its neighbors are either disconnected or at least sparsely connected. We hypothesize that preferentially vaccinating such bridge nodes would block an infection spread from a neighbor of the bridge node to an another neighbor that are otherwise not reachable to each other. We evaluate the effectiveness of using NBNC vis-a-vis degree centrality for preferential vaccination to reduce the spread of infections by conducting simulations of the spread of infections per the SIS (Susceptible-Infected-Susceptible) model on a collection of 10 complex real-world social networks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Bridge Node Centrality Tuple Infection Spread Simulations SIS Model Vaccination

来源：评论

学校读者我要写书评

暂无评论

Combining Financial Data and News Articles for Stock Price Movement Prediction Using Large Language Models

Combining Financial Data and News Articles for Stock Price M...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Elahi, Ali Taghvaei, Fatemeh University of Illinois Chicago Department of Computer Science United States University of Illinois Chicago Department of Electrical and Computer Engineering United States

ISBN: (纸本)9798350362480

Predicting financial markets and stock price movements requires analyzing a company's performance, historic price movements, industry-specific events alongside the influence of human factors such as social media and press coverage. We assume that financial reports (such as income statements, balance sheets, and cash flow statements), historical price data, and recent news articles can collectively represent aforementioned *** combine financial data in tabular format with textual news articles and employ pre-trained Large Language Models (LLMs) to predict market movements. Recent research in LLMs has demonstrated that they are able to perform both tabular and text classification tasks, making them our primary model to classify the multi-modal data. We utilize retrieval augmentation techniques to retrieve and attach relevant chunks of news articles to financial metrics related to a company and prompt the LLMs in zero, two, and four-shot settings. Our dataset contains news articles collected from different sources, historic stock price, and financial report data for 20 companies with the highest trading volume across different industries in the stock market. We utilized recently released language models for our LLM-based classifier, including GPT- 3 and 4, and LLaMA- 2 and 3 *** introduce an LLM-based classifier capable of performing classification tasks using combination of tabular (structured) and textual (unstructured) data. By using this model, we predicted the movement of a given stock's price in our dataset with a weighted F1-score of 58.5% and 59.1% and Matthews Correlation Coefficient of 0.175 for both 3-month and 6-month *** dataset and codes for this paper can be found on Github. https://***/aliielahi/FinedFMP 1 © 2024 IEEE.

关键词： Costs

来源：评论

学校读者我要写书评

暂无评论

ADVERSARIAL IMITATION LEARNING VIA BOOSTING 12

ADVERSARIAL IMITATION LEARNING VIA BOOSTING

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Chang, Jonathan D. Sreenivas, Dhruv Huang, Yingbing Brantley, Kianté Sun, Wen Department of Computer Science Cornell University United States Department of Electrical and Computer Engineering University of Illinois Urbana-Champaign United States

Adversarial imitation learning (AIL) has stood out as a dominant framework across various imitation learning (IL) applications, with Discriminator Actor Critic (DAC) (Kostrikov et al., 2019) demonstrating the effectiveness of off-policy learning algorithms in improving sample efficiency and scalability to higher-dimensional observations. Despite DAC's empirical success, the original AIL objective is on-policy and DAC's ad-hoc application of off-policy training does not guarantee successful imitation (Kostrikov et al., 2019;2020). Follow-up work such as ValueDICE (Kostrikov et al., 2020) tackles this issue by deriving a fully off-policy AIL objective. Instead in this work, we develop a novel and principled AIL algorithm via the framework of boosting. Like boosting, our new algorithm, AILBoost, maintains an ensemble of properly weighted weak learners (i.e., policies) and trains a discriminator that witnesses the maximum discrepancy between the distributions of the ensemble and the expert policy. We maintain a weighted replay buffer to represent the state-action distribution induced by the ensemble, allowing us to train discriminators using the entire data collected so far. In the weighted replay buffer, the contribution of the data from older policies are properly discounted with the weight computed based on the boosting framework. Empirically, we evaluate our algorithm on both controller state-based and pixel-based environments from the DeepMind Control Suite. AILBoost outperforms DAC on both types of environments, demonstrating the benefit of properly weighting replay buffer data for off-policy training. On state-based environments, AILBoost outperforms ValueDICE and IQ-Learn(Garg et al., 2021), achieving competitive performance with as little as one expert trajectory. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Protected Test-Time Adaptation via Online Entropy Matching: A Betting Approach 38

Protected Test-Time Adaptation via Online Entropy Matching: ...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Bar, Yarin Shaer, Shalev Romano, Yaniv Department of Computer Science Technion-Israel Institute of Technology Israel Department of Electrical and Computer Engineering Technion-Israel Institute of Technology Israel

We present a novel approach for test-time adaptation via online self-training, consisting of two components. First, we introduce a statistical framework that detects distribution shifts in the classifier's entropy values obtained on a stream of unlabeled samples. Second, we devise an online adaptation mechanism that utilizes the evidence of distribution shifts captured by the detection tool to dynamically update the classifier's parameters. The resulting adaptation process drives the distribution of test entropy values obtained from the self-trained classifier to match those of the source domain, building invariance to distribution shifts. This approach departs from the conventional self-training method, which focuses on minimizing the classifier's entropy. Our approach combines concepts in betting martingales and online learning to form a detection tool capable of quickly reacting to distribution shifts. We then reveal a tight relation between our adaptation scheme and optimal transport, which forms the basis of our novel self-supervised loss. Experimental results demonstrate that our approach improves test-time accuracy under distribution shifts while maintaining accuracy and calibration in their absence, outperforming leading entropy minimization methods across various scenarios. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Transformers Provably Learn Sparse Token Selection While Fully-Connected Nets Cannot 41

Transformers Provably Learn Sparse Token Selection While Ful...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Wang, Zixuan Wei, Stanley Hsu, Daniel Lee, Jason D. Department of Electrical and Computer Engineering Princeton University NJ United States Department of Computer Science Columbia University NY United States

The transformer architecture has prevailed in various deep learning settings due to its exceptional capabilities to select and compose structural information. Motivated by these capabilities, Sanford et al. (2023) proposed the sparse token selection task, in which transformers excel while fully-connected networks (FCNs) fail in the worst case. Building upon that, we strengthen the FCN lower bound to an average-case setting and establish an algorithmic separation of transformers over FCNs. Specifically, a one-layer transformer trained with gradient descent provably learns the sparse token selection task and, surprisingly, exhibits strong out-of-distribution length generalization. We provide empirical simulations to justify our theoretical findings. Copyright 2024 by the author(s)

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：