检索结果-内蒙古大学图书馆

41st International Conference on machine Learning, ICML 2024

作者： Li, Feiran Xu, Qianqian Bao, Shilong Yang, Zhiyong Cong, Runmin Cao, Xiaochun Huang, Qingming Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Institute of Information Science Beijing Jiaotong University Beijing China School of Control Science and Engineering Shandong University Jinan China Key Laboratory of Machine Intelligence and System Control Ministry of Education Jinan China School of Cyber Science and Tech. Sun Yat-sen University Shenzhen Campus China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores the size-invariance of evaluation metrics in Salient Object Detection (SOD), especially when multiple targets of diverse sizes co-exist in the same image. We observe that current metrics are size-sensitive, where larger objects are focused, and smaller ones tend to be ignored. We argue that the evaluation should be size-invariant because bias based on size is unjustified without additional semantic information. In pursuit of this, we propose a generic approach that evaluates each salient object separately and then combines the results, effectively alleviating the imbalance. We further develop an optimization framework tailored to this goal, achieving considerable improvements in detecting objects of different sizes. Theoretically, we provide evidence supporting the validity of our new metrics and present the generalization analysis of SOD. Extensive experiments demonstrate the effectiveness of our method. The code is available at https://***/Ferry-Li/SI-SOD. Copyright 2024 by the author(s)

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Energy Management Based on Safe Multi-Agent Reinforcement Learning for Smart Multi-Energy Buildings in Distribution Networks

SSRN

引用

SSRN 2023年

作者： Sun, Yiyun Zhang, Senlin Liu, Meiqin Zheng, Ronghao Dong, Shanling College of Electrical Engineering Zhejiang University Hangzhou310027 China National Key Laboratory of Industrial Control Technology Zhejiang University Hangzhou310027 China National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Xi’an Jiaotong University Xi’an710049 China

With the increasing expansion of renewable energy penetration in the power distribution network, the energy management for smart multi-energy buildings within the distribution network has received considerable attention. To address the distributed multi-task energy management optimization problem, we propose a safe constrained multi-agent deep reinforcement learning algorithm based on reward extrapolation. In this algorithm, hazard values are introduced to enhance cutting-edge non-safe Multi-Agent Reinforcement Learning (MARL) algorithms and meet safety constraints. By mimicking rule-based expert strategies, we infer the underlying intent of the expert strategy within a high-dimensional state-action space environment, allowing for the definition of the reward network. Additionally, the loss function for estimating the Q-network is redesigned during the training process, aiming to establish a progressively stable target value and guarantee effective convergence. Theoretical analysis is conducted to provide a comprehensive understanding, and case studies are performed to highlight the superior performance of our approach compared to existing methods. © 2023, The Authors. All rights reserved.

关键词： Constrained optimization

来源：评论

学校读者我要写书评

暂无评论

Energy Management Based on Safe Multi-Agent Reinforcement Learning for Smart Multi-Energy Buildings in Distribution Networks

SSRN

引用

SSRN 2024年

The rapid urbanization and increasing use of distributed renewable energy resources have imposed a significant burden on power networks. Smart multi-energy buildings with various energy storage systems and photovoltaic (PV) systems can be used to improve renewable energy integration and voltage quality. However, ensuring voltage stability in large multi-energy buildings presents challenges due to the coexistence of diverse energy sources and the fluctuating nature of renewable energy. This paper proposes a safe multi-energy management framework achieved by online decentralized execution and centralized training for large scale smart multi-energy buildings in distribution networks. The energy management problem is formulated as a safe constraint augmented Markov decision process, which is intractable by dynamic programming due to the large continuous state space. To solve this issue and improve the convergence speed and training process stability, a novel safe constrained multi-agent deep reinforcement learning algorithm based on reward extrapolation is proposed. In this algorithm, hazard values are introduced to enhance cutting-edge non-safe multi-agent reinforcement learning algorithms and meet safety constraints. A novel reward network is designed by imitating expert underlying intentions to ensure the rationality of the reward function for multi-objective tasks. Additionally, the loss function for estimating the Q-network is redesigned during training process to guarantee effective convergence. Theoretical analysis is conducted to provide convergence guarantee. Numerical case studies based on actual data are performed to validate the effectiveness and scalability of our approach, showing that smart multi-energy buildings can achieve superior energy management performance while ensuring voltage safety for distributed grid nodes. © 2024, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

引用

IEEE International Symposium on Parallel and Distributed Processing with Applications and IEEE International Conference on Ubiquitous Computing and Communications (ISPA/IUCC)

作者： Hui Guo Di Xie Jiahao Pang Jie He Caixu Xu Xuejian Zhang Minglang Chen Faculty of Arts and Humanities Macao University of Science and Technology Wuzhou China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University Wuzhou China college of Electronical and Information Engineering Wuzhou University Wuzhou China

At present, the self-similarity of images has received wide attention in many fields such as image compression. Most of the images have strong self-similarity through the human eye visual system observation. However, the special images with strong self-similarity for human eye observation such as biomicroscopy are not digitally reflected by a certain algorithm, this paper proposes a self-similarity coefficient algorithm that is applicable to special natural images such as biomicroscopy and aerial photography, which has high self-similarity coefficients for biomicroscopy and other images.

关键词： Photography Image coding Microscopy Visual systems Ubiquitous computing Biomedical imaging

来源：评论

学校读者我要写书评

暂无评论

Zhang Neurodynamics for Cholesky Decomposition of Matrix Stream Using Pseudo-Inverse with Transpose of Unknown

Zhang Neurodynamics for Cholesky Decomposition of Matrix Str...

引用

第40届中国控制会议

作者： Xiu Zeng Min Yang Jinjin Guo Yihong Ling Yunong Zhang School of Computer Science and Engineering Sun Yat-sen University School of Information and Communication Engineering Guangzhou Maritime University School of Electronics and Information Technology Sun Yat-sen University Guangdong Key Laboratory of Modern Control Technology Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education

Cholesky decomposition is a well-known decomposition for positive definite matrix. On account of its significantly fundamental roles in linear algebra and matrix theory, there are a lot of researches and applications based on it. In recent years,solving time-varying problems has been a research hotspot, but Cholesky decomposition of matrix stream（i.e., continuous timevarying matrix） in a simple, direct and effective equation-solving manner remains a challenging issue. In this paper, the problem of Cholesky decomposition of matrix stream is attempted and solved. First, with the aid of Zhang neurodynamics（ZN）, the objective equation at time-derivative level, including the time derivatives of matrix variable and its transpose, is obtained. In order to handle the objective equation with transpose of unknown, Kronecker product, vectorization technique and vectorized transpose matrix are utilized for better derivation. Thus, a ZN solution model using pseudo-inverse is proposed and numerically experimented. Finally, numerical experiment results substantiate the efficacy of the pseudo-inverse type ZN solution model.

关键词： Cholesky decomposition Matrix stream Transpose of unknown Zhang neurodynamics Vectorized transpose matrix

来源：评论

学校读者我要写书评

暂无评论

AADL and Modelica model combination and model conversion based on CPS 20

AADL and Modelica model combination and model conversion bas...

引用

Proceedings of the 2020 4th International Conference on Electronic Information technology and Computer Engineering

作者： Yifeng Zhu Zining Cao Fujun Wang Weiwei Lu College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing China College of Computer Science and Technology Collaborative Innovation Center of Novel Software Technology and Industrialization MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Nanjing University of Aeronautics and Astronautics and Science and Technology on Electro-optic Control Laboratory Luoyang China Science and Technology on Electro-optic Control Laboratory Luoyang China

ISBN: (纸本)9781450387811

Cyber-Physical System (CPS), which realizes the close integration of physical resources and information resources, is a distributed and asynchronous dynamic hybrid system running in different time and space. In this paper, we use AADL to model information system and use Modelica to model physical system. However, they are all semi-formal models such that we cannot do future work about formal verification. To overcome this defect, we convert them to formal models like automaton through conversion algorithms and rules. Due to the fact of probabilistic events and resource consumption in CPS, this paper adopts the weighted probabilistic hybrid automaton as formal model, and adds the properties of probability and weight to the behavior annex of AADL. Then we conduct the rule and algorithm of the model conversion. In order to ensure consistency of further work about model checking, the paper prove the equivalence of the two models by bisimulation. Finally, the capability of modeling of the extended modeling language of AADL and the significance of conversion to automaton is illustrated by the aircraft system.

关键词： Cyber-Physical System model conversion weighted probabilistic hybrid automaton AADL behavior annex bisimulation

来源：评论

学校读者我要写书评

暂无评论

ZND-ZeaD Models and Theoretics Including Proofs for Takagi Factorization of Complex Time-Dependent Symmetric Matrix

ZND-ZeaD Models and Theoretics Including Proofs for Takagi F...

引用

第33届中国控制与决策会议

作者： Zhuosong Fu Min Yang Jinjin Guo Jianrong Chen Yunong Zhang School of Computer Science and Engineering Sun Yat-sen University Guangdong Key Laboratory of Modern Control Technology Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education School of Electronics and Information Technology Sun Yat-sen University School of Public Health and Management Youjiang Medical University for Nationalities

Takagi factorization(TF) is a kind of matrix factorization(also termed,decomposition) *** this paper,the problem of TF for complex time-dependent symmetric matrix(simply termed,time-dependent Takagi factorization)is presented and theoretically *** simplify the tough problem,we first consider converting it into an equivalent equation ***,a continuous-time(CT) TF(CTTF) model is presented and derived theoretically,by adopting ZND(Zhang neural dynamics,or termed,Zhang neural network) method and a dimensional reduction *** order to facilitate the implementation on digital hardware platforms,the corresponding discrete-time(DT) ZND solution models(or termed,algorithms) with different discretization precisions of the CTTF model are presented and taken into theoretical ***,we present and investigate a discrete-time TF(DTTF) model,termed eight-instant DTTF(8IDTTF) model,by using a new eight-instant Zhang et al discretization(ZeaD,also termed,Zhang time discretization,ZTD) formula to discretize the CTTF ***,two other DTTF models,termed four-instant DTTF(4IDTTF) and six-instant DTTF(6IDTTF) models,are presented respectively,on the basis of two corresponding multiinstant ZeaD ***,the effective solution step-size intervals of the three DTTF models are further presented and proved.

关键词： Time-dependent Takagi factorization Dimensional reduction Zhang neural dynamics(ZND) Kronecker product Zhang et al discretization(ZeaD) formulas

来源：评论

学校读者我要写书评

暂无评论

Insight into Impact of Low-Content Zro2 Doping on Photocatalytic Performance of Gcn for Tetracycline Degradation

SSRN

引用

SSRN 2023年

作者： Jiang, Meng Xie, Jinqian Li, Zhanguo Huang, Mingxiu Yao, Yucen Sun, Wei Wang, Bei Key Laboratory of laser Technology and Optoelectronic Functional Materials of Hainan Province Key Laboratory of Functional Materials and Photochemistry of Haikou Collage of Chemistry and Chemical Engineering Hainan Normal University Haikou571158 China College of Optoelectronic Engineering Changchun University of Science and Technology Changchun130022 China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University Wuzhou543002 China

Semiconductor nanoparticle doping is an effective strategy for improving photocatalytic performance of graphitic carbon nitride. However, the specific mechanism for impact of semiconductor doping on photodegradation process remains unclear. Herein, we fabricated a ZrO2/gCN composite catalyst with a low doping level (1:32) and investigated doping effect on morphology, elemental state and photophysical process, including exciton, charge transfer and recombination behavior, etc. Smaller nanoscale morphology, larger specific surface area and stronger fluorescence emission for composite photocatalyst are confirmed by SEM, BET calculation, and PL spectra. Photoelectrochemical experiments indicate that charge transfer process is significantly improved and then bigger photogenerated current is achieved. Electron paramagnetic resonance results also show that •O2 signal is enhanced obviously with ZrO2 under light condition. Time resolved photoluminescence spectra further explain that ZrO2 doping is favorable for charge transfer with shortened exciton lifetime and unfavorable for charge transport with shortened recombination lifetime. Finally, the photodegradation performance for tetracycline with ZrO2/gCN is enhanced by 70% compared with pure gCN. © 2023, The Authors. All rights reserved.

关键词： Charge transfer

来源：评论

学校读者我要写书评

暂无评论

PADRE - Propeller Anomaly Data REpository for UAVs various rotor fault configurations

PADRE - Propeller Anomaly Data REpository for UAVs various r...

引用

International Conference on Unmanned Aircraft Systems (ICUAS)

作者： Radosław Puchalski Marek Kołodziejczak Adam Bondyra Jinjun Rao Wojciech Giernacki Institute of Robotics and Machine Intelligence Faculty of Control Robotics and Electrical Engineering Poznan University of Technology Poznań Poland Shanghai Key Laboratory of Intelligent Manufacturing and Robotics School of Mechatronic Engineering and Automation Shanghai University Shanghai China

The article presents a drone sensory database collected during flights with different types of propeller failures. Measurements from four accelerometers and four gyroscopes were collected during 20 flights with two types of faults occurring in different configurations in one, two, three or four rotors. The paper shows the architecture of the system and the procedure for acquiring and processing the data. Raw sensor outputs, pretreated data, and digitally processed signals were provided in a publicly available repository, the structure and purpose of which are discussed in the paper. The applicability and potential use of the shared data for other research are indicated. The provided repository should be helpful in developing methods for detecting and classifying faults in actuators of unmanned aerial vehicles (UAVs). It will be particularly useful for researchers working on data-driven methods. The default purpose of the dataset is to train artificial intelligence models that require large amounts of data.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Robust Tracking via Mamba-based Context-aware Token Learning

arXiv

引用

arXiv 2024年

作者： Xie, Jinxia Zhong, Bineng Liang, Qihua Li, Ning Mo, Zhiyi Song, Shuxiang Key Laboratory of Education Blockchain and Intelligent Technology Ministry of Education Guangxi Normal University Guilin541004 China Guangxi Key Lab of Multi-Source Information Mining & Security Guangxi Normal University Guilin541004 China Guangxi Key Laboratory of Machine Vision and Intelligent Control Wuzhou University Wuzhou543002 China

How to make a good trade-off between performance and computational cost is crucial for a tracker. However, current famous methods typically focus on complicated and time-consuming learning that combining temporal and appearance information by input more and more images (or features). Consequently, these methods not only increase the model’s computational source and learning burden but also introduce much useless and potentially interfering information. To alleviate the above issues, we propose a simple yet robust tracker that separates temporal information learning from appearance modeling and extracts temporal relations from a set of representative tokens rather than several images (or features). Specifically, we introduce one track token for each frame to collect the target’s appearance information in the backbone. Then, we design a mamba-based Temporal Module for track tokens to be aware of context by interacting with other track tokens within a sliding window. This module consists of a mamba layer with autoregressive characteristic and a cross-attention layer with strong global perception ability, ensuring sufficient interaction for track tokens to perceive the appearance changes and movement trends of the target. Finally, track tokens serve as a guidance to adjust the appearance feature for the final prediction in the head. Experiments show our method is effective and achieves competitive performance on multiple benchmarks at a real-time speed. Code and trained models will be available at https://***/GXNU-ZhongLab/TemTrack . Copyright © 2024, The Authors. All rights reserved.

关键词：

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：