检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Lu, Jianfeng Zhang, Ying Jia, Riheng Cao, Shuqin Liu, Jing Fu, Hao School of Computer Science and Technology Wuhan University of Science and Technology China Key Laboratory of Social Computing and Cognitive Intelligence Dalian University of Technology Ministry of Education China School of Computer Science and Technology Zhejiang Normal University China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology China

Federated Learning (FL) mitigates privacy leakage in decentralized machine learning by allowing multiple clients to train collaboratively locally. However, dynamic mobile networks with high mobility, intermittent connectivity, and bandwidth limitation severely hinder model updates to the cloud server. Although previous studies have typically addressed user mobility issue through task reassignment or predictive modeling, frequent migrations may result in high communication overhead. Addressing this challenge involves not only dealing with resource constraints, but also finding ways to mitigate the challenges posed by user migrations. We therefore propose an intertemporal incentive framework, FedCross, which ensures the continuity of FL tasks by migrating interrupted training tasks to feasible mobile devices. Specifically, FedCross comprises two distinct stages. In Stage 1, we address the task allocation problem across regions under resource constraints by employing a multi-objective migration algorithm to quantify the optimal task receivers. Moreover, we adopt evolutionary game theory to capture the dynamic decision-making of users, forecasting the evolution of user proportions across different regions to mitigate frequent migrations. In Stage 2, we utilize a procurement auction mechanism to allocate rewards among base stations, ensuring that those providing high-quality models receive optimal compensation. This approach incentivizes sustained user participation, thereby ensuring the overall feasibility of FedCross. Finally, experimental results validate the theoretical soundness of FedCross and demonstrate its significant reduction in communication overhead. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Chiller system Modeling using PSO Optimization based NARX approach

Chiller System Modeling using PSO Optimization based NARX ap...

引用

Data Driven Control and Learning systems (DDCLS)

作者： Zilong Song Xiaoli Li Kang Wang Yang Li Faculty of Information Technology Beijing University of Technology Beijing China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing China Engineering Research Center of Digital Community Beijing University of Technology Beijing China School of International Studies Communication University of China(CUC) Beijing China

The heating, ventilation, and air conditioning(HVAC) system consumes a large amount of energy in buildings. Accurate modeling of the HVAC refrigeration room system is crucial for building temperature control and optimization of energy consumption. In this paper, Levenberg Marquardt (LM) algorithm and particle swarm optimization (PSO) algorithm are used to establish a nonlinear autoregressive neural network model (PSO-NARX) for the modeling of water chillers in HVAC systems. NARX is a model used to describe nonlinear discrete systems. At the same time, using particle swarm optimization algorithm can improve the accuracy of the prediction model. The experimental results show that the PSO-NARX model can effectively model and predict the chiller model, and its performance is better compared to traditional DNN neural networks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Dynamics Modeling with Realistic Constraints for Robust Tracking Control of Manipulator

SSRN

引用

SSRN 2022年

作者： Zuo, Guoyu Li, Mi Liu, Lu Li, Jiangen Huang, Gao Faculty of Information Technology Beijing University of Technology Beijing100124 China Beijing Key Laboratory of Computing Intelligence and Intelligent Systems Beijing100124 China Beijing Advanced Innovation Center for Intelligent Robots and Systems Beijing Institute of Technology Beijing100081 China

To meet the requirements of precision assembly in industrial scenarios, an efficient, accurate and robust modeling control method has become an important research objective. This paper proposes a novel manipulator control method with three-level constraints, called the real deep Lagrangian network (Real-DeLaN). In this method, real data (also called physical data) are collected from the physical manipulator and utilized to train the Lagrangian dynamics network model to improve the migration ability from simulation to reality. The torque output of the dynamic model is corrected by real-time calculation of friction in the network. The contact force on the end effector of the manipulator is compensated based on the principle of virtual displacement work. Furthermore, a robust controller based on the dynamics model is designed, and the Lyapunov theorem is used to prove that the controlled system can achieve three different kinds of stability. The experimental results show that Real-DeLaN can better control the joints to perform trajectory tracking, reduce the friction error of the manipulator, and show better anti-interference ability. The robust controller further improves the robustness and tracking control effect of this method. Moreover, the method has higher practicability than other existing methods due to its good adaptability and migration ability. © 2022, The Authors. All rights reserved.

关键词： Manipulators

来源：评论

学校读者我要写书评

暂无评论

Attribute-Guided Pedestrian Retrieval: Bridging Person Re-ID with Internal Attribute Variability

Attribute-Guided Pedestrian Retrieval: Bridging Person Re-ID...

引用

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Yan Huang Zhang Zhang Qiang Wu Yi Zhong Liang Wang Center for Research on Intelligent Perception and Computing (CRIPAC) State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS) Chinese Academy of Sciences Institute of Automation Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences (UCAS). School of Electrical and Data Engineering University of Technology Sydney Australia School of Information and Electronics Beijing Institute of Technology China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

In various domains such as surveillance and smart retail, pedestrian retrieval, centering on person re-identification (Re-ID), plays a pivotal role. Existing Re-ID methodologies often overlook subtle internal attribute variations, which are crucial for accurately identifying individuals with changing appearances. In response, our paper introduces the Attribute-Guided Pedestrian Retrieval (AGPR) task, focusing on integrating specified attributes with query images to refine retrieval results. Although there has been progress in attribute-driven image retrieval, there remains a notable gap in effectively blending robust Re-ID models with intra-class attribute variations. To bridge this gap, we present the Attribute-Guided Transformer-based Pedestrian Retrieval (ATPR) framework. ATPR adeptly merges global ID recognition with local attribute learning, ensuring a co-hesive linkage between the two. Furthermore, to effectively handle the complexity of attribute interconnectivity, ATPR organizes attributes into distinct groups and applies both inter-group correlation and intra-group decorrelation regularizations. Our extensive experiments on a newly estab-lished benchmark using the RAP dataset [32] demonstrate the effectiveness of ATPR within the AGPR paradigm.

关键词： Couplings Pedestrians Correlation Surveillance Image retrieval Focusing Benchmark testing

来源：评论

学校读者我要写书评

暂无评论

Dynamic Fluorescent Imaging for Robotic Patch Clamp in vivo: Modeling and Control

Dynamic Fluorescent Imaging for Robotic Patch Clamp in vivo:...

引用

第42届中国控制会议

作者： Ruimin Li Jinyu Qiu Minghui Li Mingzhu Sun Xin Zhao Qili Zhao Institute of Robotics and Automatic Information System (IRAIS) Nankai University Tianjin Key Laboratory of Intelligent Robotic (tjKLIR) Nankai University Institute of Intelligence Technology and Robotic Systems Shenzhen Research Institute of Nankai University Beijing Advanced Innovation Center for Intelligent Robots and Systems Beijing Institute of Technology

Fluorescent imaging through fluorescent dye injection is the key technique for lighting the in vivo environment and localizing the target cell in it for patch clamp operation in vivo. At present, manual fluorescent imaging is usually a labor-intensive high-skill requisite work, significantly limiting the application of patch clamp operation in vivo. In this paper, a computational modeling of the diffusion process of the injected fluorescent dye is established using the commercial ANSYS Fluent software. Based on this model, the distribution of the mass fraction, proven to be linearly related to fluorescence intensity,was calculated to simulate the fluorescent field formed by fluorescent dye injection. Further, an ADRC controller was developed to generate feedback-controlled fluorescent fields. With the above model and controller, a mobile fluorescent field, and varying fluorescent field with constant mass fraction at the target position are generated to fulfill the requirement of in vivo environment observation, search and localization of the target cell in patch clamp operation, respectively. The above results prove the feasibility of ADRC-controlled dynamic fluorescent imaging in the patch clamp operation in vivo.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A knowledge service framework for fault diagnosis of low-earth orbit satellite constellation

A knowledge service framework for fault diagnosis of low-ear...

引用

IEEE International Conference on Web Services (ICWS)

作者： Fei Teng Yunlong Zhu Enming Zhang Xiaopeng Hu Qibo Sun Li Feng School of Computing and Artificial Intelligence Southwest Jiaotong University Chengdu China Engineering Research Center of Sustainable Urban Intelligent Transportation Ministry of Education China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

The rapid expansion of low-earth orbit satellite constellations poses a significant challenge for the operation and maintenance of thousands of satellites. In-orbit fault diagnosis helps minimize the cost of repairs, and ensure the overall reliability of the satellite constellation by identifying faults in their early stages. Compared to traditional fault detection methods, fault diagnosis requires knowledge to enhance very limited data and to infer cascading failures. In this paper, we propose a novel knowledge service framework that combines data-driven and knowledge-driven models to improve the efficiency and effectiveness of fault diagnosis. We implement a fault diagnosis service based on a constructed knowledge graph, which affords an assistant decision support function. The in-orbit deployment and simulation experiment verify the feasibility of the proposed framework, which shows great potential for fault diagnosis of low-earth orbit satellite constellation.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Hierarchical Exploration with An Active Subgoal Generation Strategy

Efficient Hierarchical Exploration with An Active Subgoal Ge...

引用

IEEE International Conference on Robotics and Biomimetics

作者： Xin Xu Guoyu Zuo Jiangeng Li Gao Huang Faculty of Information Technology Beijing University of Technology Beijing China Beijing Key Laboratory of Computing Intelligence and Intelligent Systems Beijing China Faculty of Information Beijing University of Technology Beijing China Beijing Advanced Innovation Center for Intelligent Robots and Systems Beijing China

ISBN: (纸本)9781665481106

Goal-conditioned hierarchical reinforcement learning (HRL) is a promising approach in long-horizon and complex tasks. It has been successfully used by decomposing tasks via subgoals. However, the existing goal-conditioned HRL methods often suffer from training inefficiency in the long-horizon tasks with sparse external rewards. For this problem, this paper proposes a novel framework that can train high-level policy efficiently with an effective sub goal generation method. The key component of the framework is an active subgoal generation strategy which consists of (a) an efficient high-level policy learning mechanism and (b) two measures of novelty and coverage for filtering effective subgoals. We also introduce the HER mechanism to the framework to solve the problem of sparse reward in the environment. Building upon the proposed framework, this paper develops an active hierarchical exploration strategy with an efficient subgoal generation method. We test our method on a variety of continuous control tasks. Experimental results demonstrate that the strategy outperforms the state-of-the-art HRL approaches.

关键词： Training Learning systems Representation learning Filtering Biomimetics Buildings Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

TRAIL: Trust-Aware Client Scheduling for Semi-Decentralized Federated Learning

arXiv

引用

arXiv 2024年

作者： Hu, Gangqiang Lu, Jianfeng Han, Jianmin Cao, Shuqin Liu, Jing Fu, Hao School of Computer Science and Technology Zhejiang Normal University China School of Computer Science and Technology Wuhan University of Science and Technology China Key Laboratory of Social Computing and Cognitive Intelligence Dalian University of Technology Ministry of Education China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan Universityof Science and Technology China

Due to the sensitivity of data, Federated Learning (FL) is employed to enable distributed machine learning while safeguarding data privacy and accommodating the requirements of various devices. However, in the context of semidecentralized FL, clients' communication and training states are dynamic. This variability arises from local training fluctuations, heterogeneous data distributions, and intermittent client participation. Most existing studies primarily focus on stable client states, neglecting the dynamic challenges inherent in real-world scenarios. To tackle this issue, we propose a TRust-Aware clIent scheduLing mechanism called TRAIL, which assesses client states and contributions, enhancing model training efficiency through selective client participation. We focus on a semi-decentralized FL framework where edge servers and clients train a shared global model using unreliable intra-cluster model aggregation and inter-cluster model consensus. First, we propose an adaptive hidden semi- Markov model to estimate clients' communication states and contributions. Next, we address a client-server association optimization problem to minimize global training loss. Using convergence analysis, we propose a greedy client scheduling algorithm. Finally, our experiments conducted on real-world datasets demonstrate that TRAIL outperforms state-of-the-art baselines, achieving an improvement of 8.7% in test accuracy and a reduction of 15.3% in training loss. © 2024, CC BY.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Knowledge Graph Enhanced Large Language Model Editing

arXiv

引用

arXiv 2024年

作者： Zhang, Mengqi Ye, Xiaotian Liu, Qiang Ren, Pengjie Wu, Shu Chen, Zhumin School of Computer Science and Technology Shandong University China School of Computer Science Beijing University of Posts and Telecommunications China Center for Research on Intelligent Perception and Computing State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China

Large language models (LLMs) are pivotal in advancing natural language processing (NLP) tasks, yet their efficacy is hampered by inaccuracies and outdated knowledge. Model editing emerges as a promising solution to address these challenges. However, existing editing methods struggle to track and incorporate changes in knowledge associated with edits, which limits the generalization ability of post-edit LLMs in processing edited knowledge. To tackle these problems, we propose a novel model editing method that leverages knowledge graphs for enhancing LLM editing, namely GLAME. Specifically, we first utilize a knowledge graph augmentation module to uncover associated knowledge that has changed due to editing, obtaining its internal representations within LLMs. This approach allows knowledge alterations within LLMs to be reflected through an external graph structure. Subsequently, we design a graph-based knowledge edit module to integrate structured knowledge into the model editing. This ensures that the updated parameters reflect not only the modifications of the edited knowledge but also the changes in other associated knowledge resulting from the editing process. Comprehensive experiments conducted on GPT-J and GPT-2 XL demonstrate that GLAME significantly improves the generalization capabilities of post-edit LLMs in employing edited knowledge. Copyright © 2024, The Authors. All rights reserved.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Innovative Discounted Optimal Control Design via Offline and Online Formulations for Affine systems

Innovative Discounted Optimal Control Design via Offline and...

引用

International Conference on Control and Robots (ICCR)

作者： Haiming Huang Ding Wang Junlong Wu Lingzhi Hu Faculty of Information Technology Beijing University of Technology Beijing China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing Beijing Institute of Artificial Intelligence Beijing University of Technology Beijing China Beijing Laboratory of Smart Environmental Protection Beijing University of Technology Beijing China

This paper develops a novel value iteration (VI) scheme and an online VI algorithm, to address the discounted optimal control problems of affine discrete-time nonlinear systems. First, we provide the derivation of the novel VI. Second, we analyze the convergence and monotonicity of the iterative value function sequence, as well as the admissibility of the iterative control. Third, based on the theory of the attraction domain and the novel VI scheme, an online VI algorithm is proposed to implement the stability analysis of the controlled system. It is worth noting that the current control during the online control stage is determined by the location of the current state. Finally, a simulation example is involved to demonstrate the performance of the developed algorithms.

关键词： Current control Optimal control Stability analysis Iterative algorithms Nonlinear systems Robots Numerical stability

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：