检索结果-内蒙古大学图书馆

38th AAAI Conference on Artificial Intelligence (AAAI) / 36th Conference on Innovative Applications of Artificial Intelligence / 14th Symposium on Educational Advances in Artificial Intelligence

作者： Zhang, Huaxin Wang, Xiang Xu, Xiaohao Qing, Zhiwu Gao, Changxin Sang, Nong Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Image Proc & Intelligent Control Wuhan Peoples R China Univ Michigan Ann Arbor MI 48109 USA

ISBN: (纸本)1577358872

Point-supervised Temporal Action Localization (PSTAL) is an emerging research direction for label-efficient learning. However, current methods mainly focus on optimizing the network either at the snippet-level or the instance-level, neglecting the inherent reliability of point annotations at both levels. In this paper, we propose a Hierarchical Reliability Propagation (HR-Pro) framework, which consists of two reliability-aware stages: Snippet-level Discrimination Learning and Instance-level Completeness Learning, both stages explore the efficient propagation of high-confidence cues in point annotations. For snippet-level learning, we introduce an online-updated memory to store reliable snippet prototypes for each class. We then employ a Reliability-aware Attention Block to capture both intra-video and inter-video dependencies of snippets, resulting in more discriminative and robust snippet representation. For instance-level learning, we propose a point-based proposal generation approach as a means of connecting snippets and instances, which produces high-confidence proposals for further optimization at the instance level. Through multi-level reliability-aware learning, we obtain more reliable confidence scores and more accurate temporal boundaries of predicted proposals. Our HRPro achieves state-of-the-art performance on multiple challenging benchmarks, including an impressive average mAP of 60.3% on THUMOS14. Notably, our HR-Pro largely surpasses all previous point-supervised methods, and even outperforms several competitive fully-supervised methods. Code will be available at https://***/pipixin321/HR-Pro.

关键词： Reliability

来源：评论

学校读者我要写书评

暂无评论

General Second-Order Consensus of Discrete-Time Multiagent Systems via Q-Learning Method

引用

IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS 2022年第3期52卷 1417-1425页

作者： Liu, Yifan Su, Housheng Huazhong Univ Sci & Technol Educ Minist China Sch Artificial Intelligence & Automat Image Proc & Intelligent Control Key Lab Wuhan 430074 Peoples R China

A consensus control issue is studied for general second-order multiagent systems via Q-learning method in this article. A novel second-order type with discrete-time dynamics is investigated to solve the corresponding consensus issues. Under fixed directed topology, in order to achieve general second-order consensus for the systems, a model-free Q-learning method is proposed, which can derive the coupling gains matrix without any information from the system dynamics. Moreover, applying the obtained coupling gains matrix, this general second-order multiagent systems can achieve consensus. Then, for undirected graphs, a similar corollary is obtained for this system to achieve general second-order consensus by means of the Q-learning method. Finally, the correctness of the new method is confirmed by several simulation examples.

关键词： Couplings Heuristic algorithms Multi-agent systems Topology Protocols Iterative methods System dynamics Discrete-time control general second-order consensus multiagent systems Q-learning method

来源：评论

学校读者我要写书评

暂无评论

Evaluate, explain, and explore the state more exactly: an improved Actor-Critic algorithm for complex environment

引用

NEURAL COMPUTING & APPLICATIONS 2023年第17期35卷 12271-12282页

作者： Zha, ZhongYi Wang, Bo Tang, XueSong Huazhong Univ Sci & Technol Artificial Intelligence & Automat Sch Key Lab Minist Educ Image Proc & Intelligent Control Wuhan 430074 Peoples R China Donghua Univ Coll Informat & Sci Shanghai Peoples R China

This paper proposes an Advanced Actor-Critic algorithm, which is improved based on the conventional Actor-Critic algorithm, to train the agent to play the complex strategy game StarCraft II. A series of advanced features have been incorporated, including the distributional advantage estimation, information entropy-based uncertainty estimation, self-confidence-based exploration, and normal constraint-based update strategy. A case study including seven StarCraft II mini-games is investigated to identify the effectiveness of the proposed approach, where the famous A3C algorithm is adopted as the comparative baseline. The results verify the superiority of the improved algorithm in accuracy and training efficacy, in complex environment with high-dimensional and hybrid state and action space.

关键词： Deep reinforcement learning Video game AI Advanced actor-critic Distributional advantage estimation Normal constraint

来源：评论

学校读者我要写书评

暂无评论

A new deep clustering method with application to customer selection for demand response program

引用

INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS 2023年第1期150卷

作者： Xiao, Jiang -Wen Xie, Yutao Fang, Hongliang Wang, Yan-Wu Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Minist Educ Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China

Demand response (DR) is regarded as a promising solution to the problem of renewable energy integration, while it remains one of the key barriers for DR to target the right customers, i.e. the customers with high potential in peak reduction. Exploring the electricity consumption behaviors of customers can help DR program practitioners evaluate customers' peak reduction potential. Clustering has been the most common approach to describe customers' consumption behaviors in the literature. However, the existing clustering methods suffer from either unscalability or poor performance when handling a large number of daily load curves. To overcome these issues, a new deep learning-based clustering method is proposed in this paper, which integrates both intra-cluster compactness and inter-cluster separation into the objective function of clustering for consideration. Based on the clustering results, a new customer selection framework is developed to identify the potential good candidates for DR programs, which takes the stability of consumption behaviors and the characteristics at peak hours into account. Case studies on a London dataset demonstrate the superiority of the proposed clustering method over classical K-means and two state-of-the-art deep clustering methods in terms of clustering validity indexes. Furthermore, the simulations performed on an actual price-based pilot show that the customers selected by the proposed customer selection framework contribute to 59 % of total peak reduction with only accounting for 23 % of population.

关键词： Demand response Smart meter Customer selection Deep clustering Intra-cluster compactness Inter-cluster separation ing Yan-Wu Wang Conceptualization Data curation Funding acquisition Declaration of Competing Interest The authors declare that have no known financial

来源：评论

学校读者我要写书评

暂无评论

Event-Triggered Robust control of Two-Time-Scale Systems Based on Differential Game

引用

INTERNATIONAL JOURNAL OF ROBUST AND NONLINEAR control 2025年第0期

作者： Yu, Tian-Tian Wang, Yan-Wu Lei, Yan Liu, Zhi-Wei Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Wuhan Peoples R China Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Minist Educ Wuhan Peoples R China Southwest Univ Sch Elect & Informat Engn Chongqing Peoples R China

This article studies the event-triggered robust control for linear two-time-scale systems (TTSSs) with external disturbance. To reduce control costs and optimize system performance, the stability problem is modeled as a two-player zero-sum differential game between the control policy and the external disturbance. Fast and slow subgames are derived by decoupling the TTSS and the composite optimal control strategy and the worst disturbance are obtained consequently. Furthermore, event-triggered control implementation mechanism is proposed with theoretical analysis using the Lyapunov stability theory, which can deal with the worst disturbance under the equilibrium of continuous control and realize the asymptotic stabilization of the system. Moreover, Zeno behavior of both the fast and slow subsystems is excluded. Simulation examples and a comparison study are conducted to illustrate the effectiveness of the proposed theoretical results.

关键词： event-triggered robust control stability two-time-scale system zero-sum differential game

来源：评论

学校读者我要写书评

暂无评论

Supervised Reconstruction for High-Dimensional Expensive Multiobjective Optimization

引用

IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE 2024年第2期8卷 1814-1827页

作者： Li, Hongbin Lin, Jianqing Chen, Qing He, Cheng Pan, Linqiang Huazhong Univ Sci & Technol Sch Elect & Elect Engn Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence Automat Key Lab Image Informat Proc & Intelligent Control Educ Minist China Wuhan 430074 Peoples R China

With the rising popularity of computationally expensive multiobjective optimization problems (EMOPs) in real-world applications, many surrogate-assisted evolutionary algorithms (SAEAs) have been proposed in the recent decade. Nevertheless, high-dimensional EMOPs remain challenging for existing SAEAs attributed to their requirement in massive fitness evaluations and complex models. We propose an SAEA with a supervised reconstruction strategy, namely SR-SAEA, for solving high-dimensional EMOPs. In SR-SAEA, we first select several well-converged reference solutions to form a set of reference vectors in the decision space. Then each candidate solution is projected onto these reference vectors, reflecting the closeness between the candidate solution and those reference solutions. Each candidate solution is then projected onto these reference vectors, generating a projection vector that reflects its proximity to the reference solutions. This allows the optimization of the high-dimensional decision vector to be approximated by optimizing the low-dimensional projection vector. Subsequently, a supervised autoencoder is employed to reconstruct the optimized low-dimensional projection vector back to the original decision space. Notably, the latency vector of the autoencoder is replaced with the projection vector for supervised reconstruction. An ablation study confirms the effectiveness of the proposed supervised reconstruction strategy. The superiority of SR-SAEA, compared with six state-of-the-art SAEAs, is validated on benchmark problems with up to 200 decision variables.

关键词： High-dimensional optimization multiobjective optimization surrogate-assisted optimization autoencoder

来源：评论

学校读者我要写书评

暂无评论

Multiobjective trajectory optimization with a cutting and padding encoding strategy for single-UAV-assisted mobile edge computing system

引用

SWARM AND EVOLUTIONARY COMPUTATION 2022年第0期75卷

作者： Lin, Jianqing Pan, Linqiang Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Image Informat Proc & Intelligent Control Minist China Wuhan 430074 Peoples R China

Mobile edge computing (MEC) is used to provide IT services environment and cloud computing capabilities at the edge of the network. As the technology of unmanned aerial vehicles (UAVs) matures, the growing attempts have been made to use UAVs to replace fixed ground stations for MEC due to their flexibility. In this work, we study the multiobjective trajectory optimization for mobile edge computing system assisted by a single UAV, where the UAV is used to provide computing services for Internet of Things (IoT) devices located on the ground. A multiobjective trajectory optimization problem is formulated, which not only needs to minimize the energy consumption of the MEC system to provide computing services to all IoT devices, but also minimize the task urgency indicator by optimizing the UAV's flight trajectory. In this problem, the number and the locations of hover points (HPs) of UAVs have been taken into consideration. To solve this problem, a multiobjective trajectory optimization algorithm with a cutting and padding encoding strategy is proposed, where the cutting and padding encoding strategy is used to help optimize the population whose individuals may have different lengths. The verification experiments are carried out on a set of instances with up to 400 IoT devices and the experimental results demonstrate the promising performance of the proposed algorithm for trajectory optimization problems in a single-UAV-assisted MEC system.

关键词： Mobile edge computing Internet of Things Unmanned aerial vehicle Multiobjective trajectory optimization

来源：评论

学校读者我要写书评

暂无评论

Towards Accurate image Matching by Exploring Redundancy Between Multiple Descriptors

引用

IEEE GEOSCIENCE AND REMOTE SENSING LETTERS 2023年 20卷

作者： Yu, Jinhong Sun, Kun Li, Kunqian Tang, Chuan Feng, Ruyi China Univ Geosci Sch Comp Sci Hubei Key Lab Intelligent Geoinformat Proc Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Key Lab Image Proc & Intelligent Control Minist Educ Wuhan 430074 Peoples R China Ocean Univ China Coll Engn Qingdao 266404 Peoples R China

Finding correspondences between a pair of images is the key ingredient for many applications such as localization and panorama. However, due to a variety of challenges between multiview images in practice, the results of using a single kind of descriptor may vary significantly across different scenes. In this letter, we treat the assignment task as a clustering problem and propose an image matching method that fuses multiple descriptors to tackle the above difficulties. First, we extract multiple descriptors at the keypoints on two images. Then, we compute a pairwise similarity matrix for each kind of descriptor. Afterward, we compute a weighted combination of these similarity matrices, and use it to build correspondences via a modified multikernel clustering module. The proposed method is tested on three public image datasets: two ground image sets and an unmanned aerial vehicle (UAV) image set. Experiments show that the proposed method can adapt to different number of descriptors. It significantly improves the matching accuracy in a variety of scenarios and downstream tasks.

关键词： Task analysis image matching Geoscience and remote sensing Fuses Cameras Autonomous aerial vehicles Sun Camera pose estimation image matching model fitting multidescriptor fusion unmanned aerial vehicle (UAV)

来源：评论

学校读者我要写书评

暂无评论

Adversarial filtering based evasion and backdoor attacks to EEG-based brain-computer interfaces

引用

INFORMATION FUSION 2024年 107卷

作者： Meng, Lubin Jiang, Xue Chen, Xiaoqing Liu, Wenzhong Luo, Hanbin Wu, Dongrui Huazhong Univ Sci & Technol Belt & Rd Joint Lab Measurement & Control Technol Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Artificial Intelligence & Automat Key Lab Image Proc & Intelligent Control Wuhan 430074 Peoples R China Huazhong Univ Sci & Technol Sch Civil & Hydraul Engn Wuhan 430074 Peoples R China

A brain-computer interface (BCI) enables direct communication between the brain and an external device. Electroencephalogram (EEG) is a common input signal for BCIs, due to its convenience and low cost. Most research on EEG-based BCIs focuses on the accurate decoding of EEG signals, while ignoring their security. Recent studies have shown that machine learning models in BCIs are vulnerable to adversarial attacks. This paper proposes adversarial filtering based evasion and backdoor attacks to EEG-based BCIs, which are very easy to implement. Experiments on three datasets from different BCI paradigms demonstrated the effectiveness of our proposed attack approaches. To our knowledge, this is the first study on adversarial filtering for EEG-based BCIs, raising a new security concern and calling for more attention on the security of BCIs.

关键词： Brain-computer interfaces Machine learning Adversarial attack Adversarial filtering

来源：评论

学校读者我要写书评

暂无评论

Recovering unknown topology in a two-layer multiplex network:One layer infers the other layer

引用

Science China(Technological Sciences) 2022年第7期65卷 1493-1505页

作者： LIU Hui SHANG ZhiCheng REN ZiYi LI Yan ZENG ZhiGang LU JunAn School of Artificial Intelligence and Automation Huazhong University of Science and TechnologyWuhan 430074China Key Laboratory of Image Processing and Intelligent Control of Education Ministry of China Wuhan 430074China School of Mathematics and Statistics Wuhan UniversityWuhan 430072China

There have been increasing interests in studying multiplex dynamical networks *** paper focuses on topology identiflcation of two-layer multiplex networks with peer-to-peer interlayer *** a two-layer network model in which different layers have different coupling patterns,we propose novel methods to recover unknown topological structure of one layer,using the information of the other layer known as a *** proposed methods make full use of the measured evolutional states of the multiplex network itself,and treat the layer with a known structure as an auxiliary layer which is designed to identify the unknown topological *** with the traditional synchronization-based identiflcation method,the proposed methods are in no need of constructing an additional auxiliary network to identify the unknown topological layer,and thus greatly reduce the cost of topology ***,numerical simulations validate the effectiveness of the proposed methods.

关键词： multiplex network synchronization topology identiflcation

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：