检索结果-内蒙古大学图书馆

On Approximate Opacity of Stochastic Control Systems

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Siyuan Yin, Xiang Dimarogonas, Dimos V. Zamani, Majid Division of Decision and Control Systems KTH Royal Institute of Technology Stockholm Sweden Department of Automation Shanghai Jiao Tong University Key Lab of System Control & Information Processing Ministry of Education Shanghai China Computer Science Department University of Colorado BoulderCO80309 United States Computer Science Department Ludwig Maximilian University of Munich Germany

This paper investigates an important class of information-flow security property called opacity for stochastic control systems. Opacity captures whether a system’s secret behavior (a subset of the system’s behavior that is considered to be critical) can be kept from outside observers. Existing works on opacity for control systems only provide a binary characterization of the system’s security level by determining whether the system is opaque or not. In this work, we introduce a quantifiable measure of opacity that considers the likelihood of satisfying opacity for stochastic control systems modeled as general Markov decision processes (gMDPs). We also propose verification methods tailored to the new notions of opacity for finite gMDPs by using value iteration techniques. Then, a new notion called approximate opacity-preserving stochastic simulation relation is proposed, which captures the distance between two systems’ behaviors in terms of preserving opacity. Based on this new system relation, we show that one can verify opacity for stochastic control systems using their abstractions (modeled as finite gMDPs). We also discuss how to construct such abstractions for a class of gMDPs under certain stability conditions. Copyright © 2024, The Authors. All rights reserved.

关键词： Stochastic control systems

DisenSemi: Semi-supervised Graph Classification via Disentangled Representation Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Yifan Luo, Xiao Chen, Chong Hua, Xian-Sheng Zhang, Ming Ju, Wei School of Information Technology & Management University of International Business and Economics Beijing100029 China Department of Computer Science University of California Los Angeles90095 United States Terminus Group Beijing100027 China School of Computer Science National Key Laboratory for Multimedia Information Processing Peking University Anker Embodied AI Lab Peking University Beijing100871 China College of Computer Science Sichuan University Chengdu610064 China

Graph classification is a critical task in numerous multimedia applications, where graphs are employed to represent diverse types of multimedia data, including images, videos, and social networks. Nevertheless, in the real world, labeled graph data is always limited or scarce. To address this issue, we focus on the semi-supervised graph classification task, which involves both supervised and unsupervised models learning from labeled and unlabeled data. In contrast to recent approaches that transfer the entire knowledge from the unsupervised model to the supervised one, we argue that an effective transfer should only retain the relevant semantics that align well with the supervised task. We introduce a novel framework termed DisenSemi in this paper, which learns disentangled representation for semi-supervised graph classification. Specifically, a disentangled graph encoder is proposed to generate factor-wise graph representations for both supervised and unsupervised models. Then we train two models via supervised objective and mutual information (MI)-based constraints respectively. To ensure the meaningful transfer of knowledge from the unsupervised encoder to the supervised one, we further define an MI-based disentangled consistency regularization between two models and identify the corresponding rationale that aligns well with the current graph classification task. Experiments conducted on various publicly available datasets demonstrate the effectiveness of our DisenSemi. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

Context-aware Multi-level Question Embedding Fusion for visual question answering

学校读者我要写书评

暂无评论

information Fusion 2024年 102卷

作者： Li, Shengdong Gong, Chen Zhu, Yuqing Luo, Chuanwen Hong, Yi Lv, Xueqiang School of Information Renmin University of China Beijing100872 China School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing210094 China Zhejiang Lab Zhejiang Hangzhou311121 China Department of Computer Science California State University Los Angeles CA90032 United States School of Information Science and Technology Beijing Forestry University Beijing100083 China Engineering Research Center for Forestry-Oriented Intelligent Information Processing National Forestry and Grassland Administration 100083 China Beijing Key Laboratory of Internet Culture and Digital Dissemination Research Beijing Information Science and Technology University 100101 China

Question model has been widely concerned as the cornerstone of constructing Visual Question Answering (VQA) models. Existing question models attempt to exploit word context to extract multi-level concepts for modeling multi-level questions. However, they still have many defects. For example, most question models utilize simple fusion methods to fuse shallow modules and extract parameter-unshared low-level concepts, leading to poor modeling of multi-level questions;although some question models use deep bidirectional Transformer encoder in external knowledge transfer and BERT for multi-level questions, their complexity is still high. To solve these issues, we propose a novel low-complex multi-level contextual question model, termed Context-aware Multi-level Question Embedding Fusion (CMQEF). We formalize its concepts and theories, deduce its modeling process, optimization process and feature extraction process, analyze its low complexity and high expressiveness, and prove that it defines a new way to solve parameter non-sharing for extracting parameter-shared multi-level concepts and optimize the tradeoff between expressiveness and complexity in question models. Extensive experiments on VQAv2 and VQA-CPv2 validate that comparing with the state-of-the-art, our CMQEF outperforms it on SANs and UpDn, reduces the language priors of SANs and UpDn, and has preferable interpretability and applicability. Our code is available at https://***/lsdruc/CMQEF-for-VQA. © 2023 The Author(s)

关键词： Embeddings

Attention in Attention for PET-CT Modality Consensus Lung Tumor Segmentation

学校读者我要写书评

暂无评论

Attention in Attention for PET-CT Modality Consensus Lung Tu...

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yuzhou Zhao Xinyu Zhou Haijing Guo Qianyu Guo Yan Zuo Shaoli Song Shuyong Gao Wenqiang Zhang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Department of Nuclear Medicine Fudan University Shanghai Cancer Center China Shanghai Engineering Research Center of AI Robotics Academy for Engineering and Technology Fudan University

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Combination of multi-modal PET-CT imaging for lung tumor segmentation is significant for clinical treatment. Existing methods have not fully considered the impact of noise in PET-CT on the multi-modal interaction. To address this, we propose a novel Attention in Attention Network (AiANet). AiANet can mutually learn multi-modal characteristics for segmentation through its cross-learning modules. Within the cross-learning module, we introduce two nested-attention blocks, namely Attention in Self-Attention (AiSA) and Attention in Cross-Attention (AiCA), for multi-scale feature enhancement and multi-modal feature interaction. Importantly, since traditional attention weights calculated solely or unilaterally based on PET or CT can be vulnerable to the inevitable noisy information, we embed a novel Attention in Attention (AiA) module into AiSA and AiCA. The AiA module can seek cross-modal consensus for attention weights to alleviate their noise. Experimental results on clinical PET-CT data of lung cancer demonstrate the superiority of our method.

关键词： Image segmentation Computed tomography Aggregates Noise Lung cancer Imaging Noise measurement Tumors

Attribute Surrogates Learning and Spectral Tokens Pooling in Transformers for Few-shot Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： He, Yangji Liang, Weihan Zhao, Dongyang Zhou, Hong-Yu Ge, Weifeng Yu, Yizhou Zhang, Wenqiang Nebula AI Group School of Computer Science Fudan University China Shanghai Key Lab of Intelligent Information Processing China Department of Computer Science The University of Hong Kong Hong Kong

This paper presents new hierarchically cascaded transformers that can improve data efficiency through attribute surrogates learning and spectral tokens pooling. Vision transformers have recently been thought of as a promising alternative to convolutional neural networks for visual recognition. But when there is no sufficient data, it gets stuck in overfitting and shows inferior performance. To improve data efficiency, we propose hierarchically cascaded transformers that exploit intrinsic image structures through spectral tokens pooling and optimize the learnable parameters through latent attribute surrogates. The intrinsic image structure is utilized to reduce the ambiguity between foreground content and background noise by spectral tokens pooling. And the attribute surrogate learning scheme is designed to benefit from the rich visual information in image-label pairs instead of simple visual concepts assigned by their labels. Our Hierarchically Cascaded Transformers, called HCTransformers, is built upon a self-supervised learning framework DINO and is tested on several popular few-shot learning benchmarks. In the inductive setting, HCTransformers surpass the DINO baseline by a large margin of 9.7% 5-way 1-shot accuracy and 9.17% 5-way 5-shot accuracy on miniImageNet, which demonstrates HCTransformers are efficient to extract discriminative features. Also, HCTransformers show clear advantages over SOTA few-shot classification methods in both 5-way 1-shot and 5-way 5-shot settings on four popular benchmark datasets, including miniImageNet, tieredImageNet, FC100, and CIFAR-FS. The trained weights and codes are available at https://***/StomachCold/HCTransformers. Copyright © 2022, The Authors. All rights reserved.

关键词： Efficiency

Balanced Contrastive Learning for Long-Tailed Visual Recognition

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zhu, Jianggang Wang, Zheng Chen, Jingjing Chen, Yi-Ping Phoebe Jiang, Yu-Gang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University China Shanghai Collaborative Innovation Center on Intelligent Visual Computing China Department of Computer Science and Information Technology La Trobe University Australia

Real-world data typically follow a long-tailed distribution, where a few majority categories occupy most of the data while most minority categories contain a limited number of samples. Classification models minimizing cross-entropy struggle to represent and classify the tail classes. Although the problem of learning unbiased classifiers has been well studied, methods for representing imbalanced data are under-explored. In this paper, we focus on representation learning for imbalanced data. Recently, supervised contrastive learning has shown promising performance on balanced data recently. However, through our theoretical analysis, we find that for long-tailed data, it fails to form a regular simplex which is an ideal geometric configuration for representation learning. To correct the optimization behavior of SCL and further improve the performance of long-tailed visual recognition, we propose a novel loss for balanced contrastive learning (BCL). Compared with SCL, we have two improvements in BCL: class-averaging, which balances the gradient contribution of negative classes;class-complement, which allows all classes to appear in every mini-batch. The proposed balanced contrastive learning (BCL) method satisfies the condition of forming a regular simplex and assists the optimization of cross-entropy. Equipped with BCL, the proposed two-branch framework can obtain a stronger feature representation and achieve competitive performance on long-tailed benchmark datasets such as CIFAR-10-LT, CIFAR-100-LT, ImageNet-LT, and iNaturalist2018. Our code is available at this URL. Copyright © 2022, The Authors. All rights reserved.

关键词： Benchmarking

Geometry-based spherical JND modeling for 360° display

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wei, Hongan Liu, Jiaqi Chen, Bo Lin, Liqun Chen, Weiling Zhao, Tiesong The Fujian Key Lab for Intelligent Processing and Wireless Transmission of Media Information College of Physics and Information Engineering Fuzhou University Fujian Fuzhou350116 China Fujian Science & Technology Innovation Laboratory for Optoelectronic Information of China China The Department of Computer Science University of Illinois at Urbana-Champaign UrbanaIL United States

360° videos have received widespread attention due to its realistic and immersive experiences for users. To date, how to accurately model the user perceptions on 360° display is still a challenging issue. In this paper, we exploit the visual characteristics of 360° projection and display and extend the popular just noticeable difference (JND) model to spherical JND (SJND). First, we propose a quantitative 2D-JND model by jointly considering spatial contrast sensitivity, luminance adaptation and texture masking effect. In particular, our model introduces an entropy-based region classification and utilizes different parameters for different types of regions for better modeling performance. Second, we extend our 2D-JND model to SJND by jointly exploiting latitude projection and field of view during 360° display. With this operation, SJND reflects both the characteristics of human vision system and the 360° display. Third, our SJND model is more consistent with user perceptions during subjective test and also shows more tolerance in distortions with fewer bit rates during 360° video compression. To further examine the effectiveness of our SJND model, we embed it in Versatile Video Coding (VVC) compression. Compared with the state-of-the-arts, our SJND-VVC framework significantly reduced the bit rate with negligible loss in visual quality. Copyright © 2023, The Authors. All rights reserved.

关键词： Textures

Learning Functional Dynamics from a Multilayer Brain Network

学校读者我要写书评

暂无评论

Learning Functional Dynamics from a Multilayer Brain Network

IEEE International Symposium on Biomedical Imaging

作者： Defu Yang Jiaqi Zhao Minghan Chen Yitian Xue Jingwen Zhou Shuai Wang Guorong Wu Wentao Zhu School of Information Science and Technology Hangzhou Normal University Hangzhou China Department of Psychiatry University of North Carolina at Chapel Hill Chapel Hill NC USA Research Center for Augmented Intelligence Zhejiang Lab Hangzhou China Department of Computer Science Wake Forest University Winston-Salem NC USA Intelligent Information Processing Laboratory Hangzhou Dianzi University Hangzhou China

ISBN: (数字)9798350313338

ISBN: (纸本)9798350313345

Mounting evidence reveals that functional brain network connectivity undergoes dynamic changes over time, with various disease-related alterations implicitly woven into these functional dynamics. However, existing computational approaches, particularly those using the multigraph model, often focus solely on temporal topological connectivity changes across windows, overlooking the underlying causal evolution patterns between windows. To fully capture the intricate functional dynamics of the brain, we propose a spatio-temporal dynamics learning framework that integrates temporal connectivity within windows and causal relationships between them into a unified multilayer network structure. Specifically, our multilayer graph embedding learning on manifolds can effectively preserve the intrinsic evolution pattern of functional dynamics. Experimental results on real-world datasets demonstrate that the incorporation of underlying causal relationships can significantly discover more disease-related functional dynamics.

关键词： Manifolds Computational modeling Nonhomogeneous media Brain modeling Weaving Diseases Biomedical imaging

ESM All-Atom: Multi-scale Protein Language Model for Unified Molecular Modeling

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zheng, Kangjie Long, Siyu Lu, Tianyu Yang, Junwei Dai, Xinyu Zhang, Ming Nie, Zaiqing Ma, Wei-Ying Zhou, Hao School of Computer Science National Key Laboratory for Multimedia Information Processing Peking University Anker Embodied AI Lab Peking University Beijing100871 China School of Artificial Intelligence National Key Laboratory for Novel Software Technology Nanjing University China Department of Computer Science Tsinghua University China Tsinghua University China PharMolix Inc

Protein language models have demonstrated significant potential in the field of protein engineering. However, current protein language models primarily operate at the residue scale, which limits their ability to provide information at the atom level. This limitation prevents us from fully exploiting the capabilities of protein language models for applications involving both proteins and small molecules. In this paper, we propose ESM-AA (ESM All-Atom), a novel approach that enables atom-scale and residue-scale unified molecular modeling. ESM-AA achieves this by pretraining on multi-scale code-switch protein sequences and utilizing a multi-scale position encoding to capture relationships among residues and atoms. Experimental results indicate that ESM-AA surpasses previous methods in protein-molecule tasks, demonstrating the full utilization of protein language models. Further investigations reveal that through unified molecular modeling, ESM-AA not only gains molecular knowledge but also retains its understanding of proteins. © 2024, CC BY-NC-ND.

关键词： Proteins