检索结果-内蒙古大学图书馆

arXiv 2025年

作者： Wei, Hao Zhou, Yanhui Jia, Yiwen Ge, Chenyang Anwar, Saeed Mian, Ajmal National Key Laboratory of Human-Machine Hybrid Augmented Intelligence Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an710049 China School of Information and Telecommunication Xi’an Jiaotong University Xi’an710049 China Department of Computer Science and Software Engineering The University of Western Australia Crawley PerthWA6009 Australia

Perceptual image compression has shown strong potential for producing visually appealing results at low bitrates, surpassing classical standards and pixel-wise distortion-oriented neural methods. However, existing methods typically improve compression performance by incorporating explicit semantic priors, such as segmentation maps and textual features, into the encoder or decoder, which increases model complexity by adding parameters and floating-point operations. This limits the model’s practicality, as image compression often occurs on resource-limited mobile devices. To alleviate this problem, we propose a lightweight perceptual Image Compression method using Implicit Semantic Priors (ICISP). We first develop an enhanced visual state space block that exploits local and global spatial dependencies to reduce redundancy. Since different frequency information contributes unequally to compression, we develop a frequency decomposition modulation block to adaptively preserve or reduce the low-frequency and high-frequency information. We establish the above blocks as the main modules of the encoder-decoder, and to further improve the perceptual quality of the reconstructed images, we develop a semantic-informed discriminator that uses implicit semantic priors from a pretrained DINOv2 encoder. Experiments on popular benchmarks show that our method achieves competitive compression performance and has significantly fewer network parameters and floating point operations than the existing state-of-the-art. We will release the code and trained models. Copyright © 2025, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Hierarchical Decompositions of Stochastic Pursuit-Evasion Games

arXiv

引用

arXiv 2022年

作者： Guan, Yue Afshari, Mohammad Zhang, Qifan Tsiotras, Panagiotis The School of Aerospace Engineering Georgia Institute of Technology AtlantaGA United States The Institute for Robotics and Intelligent Machines Georgia Institute of Technology AtlantaGA United States Software Engineer at Google Mountain ViewCA United States

In this work we present a hierarchical framework for solving discrete stochastic pursuit-evasion games (PEGs) in large grid worlds. With a partition of the grid world into superstates (e.g., "rooms"), the proposed approach creates a two-resolution decision-making process, which consists of a set of local PEGs at the original state level and an aggregated PEG at the superstate level. Having much smaller cardinality, both the local games and the aggregated game can be easily solved to a Nash equilibrium. To connect the decision-making at the two resolutions, we use the Nash values of the local PEGs as the rewards for the aggregated game. Through numerical simulations, we show that the proposed hierarchical framework significantly reduces the computation overhead, while still maintaining a satisfactory level of performance when competing against the flat Nash policies. © 2022, CC BY.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Negotiated Reasoning: On Provably Addressing Relative Over-Generalization

arXiv

引用

arXiv 2023年

作者： Sheng, Junjie Li, Wenhao Jin, Bo Zha, Hongyuan Wang, Jun Wang, Xiangfeng School of Computer Science and Technology East China Normal University Shanghai China School of Data Science The Chinese University of Hong Kong Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of Software Engineering Tongji University Shanghai China

Over-generalization is a thorny issue in cognitive science, where people may become overly cautious due to past experiences. Agents in multi-agent reinforcement learning (MARL) also have been found suffering relative over-generalization (RO) as people do and stuck to sub-optimal cooperation. Recent methods have shown that assigning reasoning ability to agent can mitigate RO algorithmically and empirically, but there has been a lack of theoretical understanding of RO, let alone designing provably RO-free methods. This paper first proves that RO can be avoided when the MARL method satisfies a consistent reasoning requirement under certain conditions. Then we introduce a novel reasoning framework, called negotiated reasoning, that first builds the connection between reasoning and RO with theoretical justifications. After that, we propose an instantiated algorithm, Stein variational negotiated reasoning (SVNR), which uses Stein variational gradient descent to derive a negotiation policy that provably avoids RO in MARL under maximum entropy policy iteration. The method is further parameterized with neural networks for amortized learning, making computation efficient. Numerical experiments on many RO-challenged environments demonstrate the superiority and efficiency of SVNR compared to state-of-the-art methods in addressing RO. © 2023, CC BY.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Cyberbullying Classification Using a Natural Language Processing and Machine Learning Techniques

Cyberbullying Classification Using a Natural Language Proces...

引用

Advances in Computing Research on Science engineering and Technology (ACROSET), International Conference on

作者： Nikhilesh Pramod Mankar Tushar Mahure Anushka A. Joshi Sagar Dhanraj Pande Aditya Khamparia Ankita Shrikrushna Nathe Fadi Al Turjman School of Engineering and Technology Pimpri Chinchwad University (PCU) Pune India Independent Researcher Amravati India Department of Computer Science Babasaheb Bhimrao Ambedkar University Amethi India Department of Information Technology Kavikulguru Institute of Technology and Science Nagpur India Artificial Intelligence Software Information Systems Engineering Departments AI and Robotics Institute Near East University Nicosia Turkey Research Center for AI and IoT Faculty of Engineering University of Kyrenia Kyrenia Turkey

ISBN: (数字)9798350388800

ISBN: (纸本)9798350388817

The usage of online entertainment has increased dramatically after some time with the enhancement of the Internet and has turned into the most compelling systems administration stage in this century. Notwithstanding, the upgrade of a social network regularly makes adverse consequences on society that add to two or three terrible peculiarities, for example, online maltreatment, badgering cyberbullying, cybercrime, and internet savaging. Cyberbullying habitually prompts genuine mental and actual misery, especially for ladies and kids, and even occasionally make them to use self-destruction methods. Multiple number of episodes have as of late happened worldwide because of online badgering, like sharing private talks, bits of hearsay, and sexual comments. In this way, the distinguishing proof of harassing messages or messages via online entertainment has gained a significant measure for development for consideration with in scientists. The ground or motive for this exploration is to create a full-fledged way or a method to differentiate online harmful and tormenting drafts by combining various methods and techniques in the domain of natural language processing and AI

关键词： Cyberbullying Entertainment industry Machine learning Gain measurement Natural language processing Internet Bayes methods Computer crime

来源：评论

学校读者我要写书评

暂无评论

A Knowledge Distillation Based Deep Learning Framework for Cropped Images Detection in Spatial Domain

SSRN

引用

SSRN 2022年

作者： Hussain, Israr Tan, Shunquan Huang, Jiwu College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Shenzhen518060 China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen University Shenzhen518000 China

Cropping an image is a typical image editing technique that seeks to locate viewpoints with appropriate image composition. It is also a commonly used post-processing technique to reduce the traces in a maliciously tempered image. The automated detection of cropped images based on inherent blocking artifacts introducing in cropped images is considered a challenging task in the field of Multimedia forensics. Deep neural networks achieve state-of-the-art performance in this field with high computational complexity and substantial storage needs. To design a network which can obtain better performance with less number of network parameters is always an active area of research in the field of computer vision. To address this problem, a lightweight knowledge distillation based deep learning framework for cropped images in spatial domain is proposed in this paper. We used Efficient-Net-B0 pre-trained in ImageNet as a teacher model, which makes it more robust and faster to converge in this down stream task. The model was trained on 20,000 cropped and uncropped images from our own constructed dataset. We then applied its knowledge to a more compact model called the student model. Finally, the best performing lightweight model is chosen as the final prediction model, with a testing accuracy of 98.44% on the test dataset, which is superior than the other methods. Our experimental results demonstrates that it is possible to combine the advantages of both large and small models in order to achieve the goals of minimal resource consumption and high accuracy. © 2022, The Authors. All rights reserved.

关键词： Discrete cosine transforms

来源：评论

学校读者我要写书评

暂无评论

Two-Stream (2+1)D CNN Based on Frame Difference Attention for Driver Behavior Recognition

Two-Stream (2+1)D CNN Based on Frame Difference Attention fo...

引用

International Conference on Dependable Systems and Their Applications (DSA)

作者： Yang Zou Shaoyi Du Hongcheng Han Yuying Liu Zhiqiang Tian National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an China Shunan Academy of Artificial Intelligence Ningbo China School of Software Engineering Xi'an Jiaotong University Xi'an China

Driving behavior recognition is a key technology in human-machine cooperative driving, which plays an important role in the construction of an intelligent transportation environment. Due to the similarity of drivers ' actions and backgrounds in driving environments, it's challenging for existing behavior recognition methods to recognize behaviors in driving environments precisely. To solve these problems, we propose the two-stream (2+1)D CNN with Frame Difference Attention Module. First, we use two streams to process the information captured from the face camera and road camera and use the Squeeze and Excitation attention module to fuse the stream information. Second, we propose the Frame Difference Attention Module to enhance the changes between adjacent frames so that the subtle actions in driving environments can be captured easier. We test our network on the Brain4Cars dataset and have a good performance on it.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Explicitly fusing plug-and-play guidance of source prototype into target subspace for domain adaptation

引用

Information Fusion 2025年 123卷

作者： Luo, Hao Tian, Zhiqiang Jiao, Panpan Liu, Meiqin Du, Shaoyi Nan, Kai School of Software Engineering Xi'an Jiaotong University Shaanxi Xi'an710049 China State Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications and Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Shaanxi Xi'an710049 China Department of Joint Surgery Honghui Hospital Xi'an Jiaotong University Shaanxi Xi'an710054 China

The commonly used maximum mean discrepancy (MMD) criterion has two main drawbacks when reducing cross-domain distribution gaps: firstly, it reduces the distribution discrepancy in a global manner, potentially ignoring local structural information between domains, and secondly, its performance heavily relies on the often-unstable pseudo-label refinement process. To solve these problems, we introduce two universal plug-and-play modules: dynamic prototype pursuit (DPP) regularization and bi-branch self-training (BST) mechanism. Firstly, DPP introduces a new inter-class perspective to stabilize MMD by assigning a source prototype to each target sample. This allows us to utilize inter-class data structure information for better alignment. Next, BST is a novel non-parametric pseudo-label refinement mechanism that updates pseudo labels of target data using a classifier trained on the same distribution as the target domain. This avoids the distribution gap issue, making BST more likely to generate accurate target pseudo labels. Importantly, DPP and BST are universal plug-and-play modules for shallow domain adaptation methods. To demonstrate this, experiments of 3 MMD-based models incorporated with DPP and BST are conducted on Office-Caltech, Reuters21578, and Berlin-Emovo-Tess datasets. Experimental results show that these models incorporated with DPP and BST generally achieve better results compared to not using DPP and BST in terms of multiple metrics including accuracy, F1-score, MCC, and false positive rates. Code of 3 different DA methods enhanced by the plug-and-play DPP and BST is available at: https://***/Evelhz/DPP-and-BST. © 2025 Elsevier B.V.

关键词： Bi-branch self-training Dynamic prototype pursuit Shallow domain adaptation Subspace learning

来源：评论

学校读者我要写书评

暂无评论

F2A2: flexible fully-decentralized approximate actor-critic for cooperative multi-agent reinforcement learning

The Journal of Machine Learning Research

引用

The Journal of Machine Learning Research 2023年第1期24卷 8447-8521页

作者： Wenhao Li Bo Jin Xiangfeng Wang Junchi Yan Hongyuan Zha School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen China School of Software Engineering Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai China School of Computer Science and Technology Key Laboratory of Mathematics and Engineering Applications Ministry of Education East China Normal University Shanghai China Department of Computer Science and Engineering Key Laboratory of Artificial Intelligence Ministry of Education Shanghai Jiao Tong University Shanghai China

Traditional centralized multi-agent reinforcement learning (MARL) algorithms are sometimes unpractical in complicated applications due to non-interactivity between agents, the curse of dimensionality, and computation complexity. Hence, several decentralized MARL algorithms are motivated. However, existing decentralized methods only handle the fully cooperative setting where massive information needs to be transmitted in training. The block coordinate gradient descent scheme they used for successive independent actor and critic steps can simplify the calculation, but it causes serious bias. This paper proposes a exible fully decentralized actor-critic MARL framework, which can combine most of the actor-critic methods and handle large-scale general cooperative multi-agent settings. A primal-dual hybrid gradient descent type algorithm framework is designed to learn individual agents separately for decentralization. From the perspective of each agent, policy improvement and value evaluation are jointly optimized, which can stabilize multi-agent policy learning. Furthermore, the proposed framework can achieve scalability and stability for the large-scale environment. This framework also reduces information transmission by the parameter sharing mechanism and novel modeling-other-agents methods based on theory-of-mind and online supervised learning. Sufficient experiments in cooperative Multi-agent Particle Environment and StarCraft II show that the proposed decentralized MARL instantiation algorithms perform competitively against conventional centralized and decentralized methods.

关键词： cooperative MARL decentralized actor-critic primal-dual method

来源：评论

学校读者我要写书评

暂无评论

GSA-Gaze: Generative Self-adversarial Learning for Domain Generalized Driver Gaze Estimation

GSA-Gaze: Generative Self-adversarial Learning for Domain Ge...

引用

International Conference on Intelligent Transportation

作者： Hongcheng Han Zhiqiang Tian Yuying Liu Shengpeng Li Dong Zhang Shaoyi Du National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi'an Jiaotong University Xi'an China Faculty of Electronic and Information Engineering School of Software Engineering Xi'an Jiaotong University Xi'an China Faculty of Electronic and Information Engineering School of Automation Science and Engineering Xi'an Jiaotong University Xi'an China

Estimating driver gaze accurately is critical for the human-machine cooperative driving, but the significant facial appearance diversions caused by background, illumination, personal characteristics, etc. pose a challenge to the generalizability of gaze estimation models. In this paper, we propose the generative self-adversarial learning mechanism for generalized gaze estimation that aims to learn general gaze features while eliminating sample-specific features and preventing cross-domain feature over-fitting. Firstly, to reduce information redundancy, the feature encoder is designed based on pyramid-grouped convolution to extract a sparse feature representation from the facial appearance. Secondly, the gaze regression module supervises the model to learn as many gaze-relevant features as possible. Thirdly, the adversarial image reconstruction task prompts the model to eliminate the domain-specific features. The adversarial learning of the gaze regression and the image reconstruction tasks guides the model to learn only general gaze features across domains, preventing cross-domain feature over-fitting, enhancing the domain generalization capability. The results of cross-domain testing of four active gaze datasets prove the effectiveness of the proposed method. The code is available at https://***/HongchengHan/GSA-Gaze

关键词：

来源：评论

学校读者我要写书评

暂无评论

Spectrogram-based Aging Assessment for Insulations on Stator Coil Bar

Spectrogram-based Aging Assessment for Insulations on Stator...

引用

2021 China Automation Congress, CAC 2021

作者： Liu, Shuo Zhang, Yue Hu, Bo Ding, Jian Wang, Jianji Institute of Artificial Intelligence and Robotics College of Artificial Intelligence Xi'an Jiaotong University Xi'an710049 China School of Software Engineering Xi'an Jiaotong University Xi'an710049 China DongFang Electric Machinery Co. Ltd Deyang618000 China

ISBN: (纸本)9781665426473

Based on four kinds of spectrograms, this paper studies how to effectively predict the age of insultations on stator coil bar. The prediction of using time can guide the assessment of stator coil bar. In this work, some samples with different ages and temperatures of epoxy insulations are generated in an accelerated aging test. Then the microscopic morphology spectrograms are obtained for the samples. We use the histogram of oriented gradient (HOG) and gray-level co-occurrence matrix (GLCM) to extract the features of the spectrograms. Finally, support vector machine (SVM) is used to classify and predict the using time of samples based on HOG and GLCM features. This method with accuracy 90.3333% on the classification and prediction for the age of insulations on stator coil bar. © 2021 IEEE

关键词： Support vector machines

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：