检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Chen, Dingshuo Li, Zhixun Ni, Yuyan Zhang, Guibin Wang, Ding Liu, Qiang Wu, Shu Yu, Jeffrey Xu Wang, Liang New Laboratory of Pattern Recognition State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Hong Kong Academy of Mathematics and Systems Science Chinese Academy of Sciences China Tongji University China

With the emergence of various molecular tasks and massive datasets, how to perform efficient training has become an urgent yet under-explored issue in the area. Data pruning (DP), as an oft-stated approach to saving training burdens, filters out less influential samples to form a coreset for training. However, the increasing reliance on pretrained models for molecular tasks renders traditional in-domain DP methods incompatible. Therefore, we propose a Molecular data Pruning framework for enhanced Generalization (MolPeg), which focuses on the source-free data pruning scenario, where data pruning is applied with pretrained models. By maintaining two models with different updating paces during training, we introduce a novel scoring function to measure the informativeness of samples based on the loss discrepancy. As a plug-and-play framework, MolPeg realizes the perception of both source and target domain and consistently outperforms existing DP methods across four downstream tasks. Remarkably, it can surpass the performance obtained from full-dataset training, even when pruning up to 60-70% of the data on HIV and PCBA dataset. Our work suggests that the discovery of effective data-pruning metrics could provide a viable path to both enhanced efficiency and superior generalization in transfer learning. © 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

Extension Removal in Abstract Argumentation: Necessary and Sufficient Conditions

Extension Removal in Abstract Argumentation: Necessary and S...

引用

2023 IEEE International Conference on Agents, ICA 2023

作者： Zhang, Huan Zhang, Songmao Academy of Mathematics and Systems Science Chinese Academy of Sciences Key Laboratory of Madis Beijing China School of Mathematical Sciences University of Chinese Academy of Sciences Beijing China

ISBN: (纸本)9798350312782

In the research on dynamics of abstract argumentation frameworks, Baumann et al. proposed the extension removal problem for the first time in 2019 and provided an axiomatization of removal operators. In contrast, we are interested in how to solve the extension removal problem in practice. Our consideration includes designing a special removal operator for preventing undesired extensions from being accepted whatsoever in the modified AF;and limiting AF modification to normal expansion so as to honor the original arguments and their attacks. The contribution of this paper consists of four jointly necessary and sufficient conditions identified for solving the extension removal problem under preferred, stable, semi-stable, grounded, ideal and complete semantics, as well as an impossibility conclusion under naive semantics. Such conditions are both minimum and absolute which can be used to determine when the removal problem is (un)solvable. © 2023 IEEE.

关键词： Expansion

来源：评论

学校读者我要写书评

暂无评论

THE OPTIMAL STRATEGY AGAINST HEDGE ALGORITHM IN REPEATED GAMES

arXiv

引用

arXiv 2023年

作者： Guo, Xinxiang Mu, Yifen The Key Laboratory of Systems and Control Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing China

This paper aims to solve the optimal strategy against a well-known adaptive algorithm, the Hedge algorithm, in a finitely repeated 2 × 2 zero-sum game. In the literature, related theoretical results are very rare. To this end, we make the evolution analysis for the resulting dynamical game system and build the action recurrence relation based on the Bellman optimality equation. First, we define the state and the State Transition Triangle Graph (STTG);then, we prove that the game system will behave in a periodic-like way when the opponent adopts the myopic best response. Further, based on the myopic path and the recurrence relation between the optimal actions at time-adjacent states, we can solve the optimal strategy of the opponent, which is proved to be periodic on the time interval truncated by a tiny segment and has the same period as the myopic path. Results in this paper are rigorous and inspiring, and the method might help solve the optimal strategy for general games and general *** Codes 91A05, 91A20 Copyright © 2023, The Authors. All rights reserved.

关键词： Adaptive algorithms

来源：评论

学校读者我要写书评

暂无评论

Cross-Dimensional mathematics: A Mathematical Foundation for STP/STA

arXiv

引用

arXiv 2024年

作者： Cheng, Daizhan Key Laboratory of Systems and Control Academy of Mathematics and Systems Sciences Chinese Academy of Sciences Beijing100190 China

A new mathematical structure, called the cross-dimensional mathematics (CDM), is proposed. The CDM considered in this paper consists of three parts: hyper algebra, hyper geometry, and hyper Lie group/Lie algebra. Hyper algebra proposes some new algebraic structures such as hyper group, hyper ring, and hyper module over matrices and vectors with mixed dimensions (MVMDs). They have sets of classical groups, rings, and modules as their components and cross-dimensional connections among their components. Their basic properties are investigated. Hyper geometry starts from mixed dimensional Euclidian space, and hyper vector space. Then the hyper topological vector space, hyper inner product space, and hyper manifold are constructed. They have a joined cross-dimensional geometric structure. Finally, hyper metric space, topological hyper group and hyper Lie algebra are built gradually, and finally, the corresponding hyper Lie group is introduced. All these concepts are built over MVMDs, and to reach our purpose in addition to existing semi-tensor products (STPs) and semi-tensor additions (STAs), a couple of most general STP and STA are introduced. Some existing structures/results about STPs/STAs have also been resumed and integrated into this CDM. © 2024, CC BY.

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Deep Co-Training Partial Least Squares Model for Semi-Supervised Industrial Soft Sensing

引用

IEEE Transactions on systems, Man, and Cybernetics: systems 2025年第5期55卷 3363-3371页

作者： Zheng, Junhua Zhou, Le Ye, Lingjian Ge, Zhiqiang Zhejiang University of Science and Technology School of Automation and Electrical Engineering Hangzhou310023 China Huzhou University Huzhou Key Laboratory of Intelligent Sensing and Optimal Control for Industrial Systems School of Engineering Huzhou313000 China Southeast University School of Mathematics Nanjing210096 China

Data-driven soft sensing has become quite popular in recent years, which can provide real-time estimations of key variables in industrial processes. While the introduction of deep learning does improve the prediction performance, it is highly restricted to the number of labeled training data, as well as large computational burden and cumbersome parameter tuning procedures. How to break through the bottleneck of data-drive models in terms of limited labeled data and high computational complexity should be one of the main recent focuses in the field of industrial soft sensing. In this article, a deep co-training PLS (deep CT-PLS) model is proposed to extend the ordinary PLS model to the semi-supervised deep form. While the deep model can efficiently extract inherent natures of process data, the co-training strategy makes lots of unlabeled data useful through a two-view cross training and annotation process. In this case, the performance restriction of the deep PLS model can be greatly relieved, with the incorporation of additional unlabeled data, while at the same time the designed model structure keeps in a low computational complexity. Based on the case study on a real industrial production process, the deep CT-PLS model can significantly improve the soft sensing performance. © 2013 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Semantic Coherence Guided Multiview Similarity for Image Classification with Varied Supervision

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第6期5卷 3072-3082页

作者： Zhang, Chunjie Cao, Ping Bai, Huihui Zheng, Xiaolong Beijing Jiaotong University Institute of Information Science Beijing100044 China Beijing Jiaotong University Beijing Key Laboratory of Advanced Information Science and Network Technology Beijing100044 China University of Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems The State of Key Laboratory of Management and Control for Complex System Institute of Automation Beijing100190 China University of Chinese Academy of Sciences School of Artificial Intelligence Beijing100190 China

Pairwise similarity has been widely used for image classification by propagating the class information from labeled images to unlabeled images and predicting the classes of unlabeled images accordingly. Although widely used, pairwise similarity based classification methods have two drawbacks. On one hand, visual-semantic similarity consistency does not always hold due to semantic gap. On the other hand, the reliability of pairwise similarity is also influenced by the number and underlying distribution of labeled images. A well-designed classification model cannot be easily generalized to few-shot classification tasks. To solve these two problems, in this article, we propose a semantic coherence guided multiview similarity (SCG-MV) for image classification method with varied supervision levels. We measure the semantic coherence between image pairs for classifier learning. This is achieved by first conducting line search along the line segment between each image pair, and then measure the semantics of each interval. The similarity of one image pair can then be obtained by aggregating the semantic coherence of all the intervals. In this way, we can model image similarities more finely and discriminatively. To make use of multiview information, we also use the semantic coherence guided image similarities to learn multiview classifiers. The proposed semantic coherence guided method can be combined with other similarity based methods for image classification with various levels of supervision. We conduct both few-shot and fully supervised classification experiments on four datasets with the results well prove the discriminative power and generalization ability of the proposed method. © 2020 IEEE.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Signal Processing via Cross-Dimensional Projection

arXiv

引用

arXiv 2024年

作者： Cheng, Daizhan Key Laboratory of Systems and Control Academy of Mathematics and Systems Sciences Chinese Academy of Sciences Beijing100190 China

Using projection between Euclidian spaces of different dimensions, the signal compression and decompression become straightforward. This encoding/decoding technique requires no preassigned measuring matrix as in compressed sensing. Moreover, in application there is no dimension or size restrictions. General formulas for encoding/decoding of any finite dimensional signals are provided. Their main properties are revealed. Particularly, it is shown that under the equivalence assumption the technique provides the best approximation with least square error. © 2024, CC BY.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

RNN learning controller trained by ADRC 41

RNN learning controller trained by ADRC

引用

第41届中国控制会议

作者： Yutao Zhao Mengyang Wang Yi Huang Linfeng Li Yongchun Xie Key Lab.of System and Control Academy of Mathematics and Systems ScienceChinese Academy of Sciences School of Mathematical Science University of Chinese Academy of Sciences Beijing Institute of Control Engineering

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

In this paper,the problem of training a recurrent neural network(RNN) controller to approximate an active disturbance rejection control(ADRC) is *** learning methods,online learning and offline learning,are *** the online setting,the RNN learning controller is trained by the error of the control quantity compared to that of ADRC in real time,while for the offline setting,the RNN controller is trained by the error of the whole state trajectory compared to that of ADRC obtained in *** randomized weight initialization,success rates of trained online and offline RNN learning controllers are compared by ***,the robustness property of the trained RNN controller is analyzed.A metric of robustness,which can be calculated from the weights of the trained RNN controller is also proposed.

关键词： Recurrent Neural Network(RNN) learning controller Active disturbance rejection control(ADRC) Online learning Offline learning Robustness

来源：评论

学校读者我要写书评

暂无评论

Towards Intelligent and Trustworthy Crowdsourcing systems with Cyber-physical Social Intelligence via Retrieval-augmented Generation

Towards Intelligent and Trustworthy Crowdsourcing Systems wi...

引用

2024 International Conference on Cyber-Physical Social Intelligence, ICCSI 2024

作者： Yang, Jing Wang, Xingxia Ni, Qinghua Liu, Yuhang Wang, Jing Wang, Fei-Yue Chinese Academy of Sciences State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Beijing100190 China Macau University of Science and Technology Faculty of Innovation Engineering 999078 China Macau University of Science and Technology 999078 China Chinese Academy of Sciences State Key Laboratory for Management and Control of Complex Systems Beijing100190 China

ISBN: (纸本)9798350376739

In recent years, crowdsourcing systems that tackle complex tasks through the collective efforts of many individuals have garnered substantial attention. However, the existing crowd-sourcing systems face some challenges such as overreliance on human experience and effort, information overload, and risks of privacy breaches. To overcome these challenges, this paper proposes an intelligent and trustworthy crowdsourcing frame-work based on blockchain and retrieval-augmented generation. The framework centers around foundational models to integrate the three distinct and sequentially interdependent phases: task decomposition, task allocation or recommendation, and task execution, enabling the automatic provision of a personalized, comfortable, and efficient user experience. Blockchain and smart contracts are utilized to protect user privacy and interests without leaning on any third trusted authority. Additionally, the operational process of the framework is explained and validated by taking printed circuit board design and manufacturing as a case study and ChatGPT as its generator. © 2024 IEEE.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

Detecting and Exploiting Unit Commitment Structures in MIPs 6

Detecting and Exploiting Unit Commitment Structures in MIPs

引用

6th International Conference on Power and Energy Technology, ICPET 2024

作者： Su, Zhao-Gang Chen, Sheng-Jie Li, Yu-Fei Chen, Liang Kou, Caixia Cai, Zhi Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing China Institute of Applied Mathematics Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing China College of Sciences Beijing University of Posts and Telecommunications Beijing China Beijing Key Laboratory of Research and System Evaluation of Power Dispatching Automation Technology China Electric Power Research Institute Beijing China

ISBN: (纸本)9798350375855

The unit commitment (UC) problem has been extensively researched in the literature, which is typically formulated as a mixed integer programming (MIP) problem. However, current studies lack effective methods to identify potential UC problem structures within a general MIP problem, making it challenging to leverage well-established valid inequalities for UC problems to accelerate the solution process. In this paper, we present a novel detection method that automatically identifies UC structures within an MIP problem by analyzing the relationships between variables and constraints. Moreover, we enhance the MIP formulation by incorporating three well-studied classes of valid inequalities as cutting planes, based on the identified UC structures. Our experiments demonstrate that our proposed UC detection method greatly improves computational efficacy of a standard MIP solver compared to the default setting. © 2024 IEEE.

关键词： Mixed-integer linear programming

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：