检索结果-内蒙古大学图书馆

Clean-label Backdoor Attacks on Video Recognition Models

学校读者我要写书评

暂无评论

Clean-Label Backdoor Attacks on Video Recognition Models

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Shihao Zhao Xingjun Ma Xiang Zheng James Bailey Jingjing Chen Yu-Gang Jiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University School of Computing and Information Systems The University of Melbourne

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

Deep neural networks (DNNs) are vulnerable to backdoor attacks which can hide backdoor triggers in DNNs by poisoning training data. A backdoored model behaves normally on clean test images, yet consistently predicts a particular target class for any test examples that contain the trigger pattern. As such, backdoor attacks are hard to detect, and have raised severe security concerns in real-world applications. Thus far, backdoor research has mostly been conducted in the image domain with image classification models. In this paper, we show that existing image backdoor attacks are far less effective on videos, and outline 4 strict conditions where existing attacks are likely to fail: 1) scenarios with more input dimensions (eg. videos), 2) scenarios with high resolution, 3) scenarios with a large number of classes and few examples per class (a ``sparse dataset"), and 4) attacks with access to correct labels (eg. clean-label attacks). We propose the use of a universal adversarial trigger as the backdoor trigger to attack video recognition models, a situation where backdoor attacks are likely to be challenged by the above 4 strict conditions. We show on benchmark video datasets that our proposed backdoor attack can manipulate state-of-the-art video models with high success rates by poisoning only a small proportion of training data (without changing the labels). We also show that our proposed backdoor attack is resistant to state-of-the-art backdoor defense/detection methods, and can even be applied to improve image backdoor attacks. Our proposed video backdoor attack not only serves as a strong baseline for improving the robustness of video models, but also provides a new perspective for more understanding more powerful backdoor attacks.

关键词： Training Data models Toxicology Perturbation methods Training data Image resolution Pipelines

Efficient Video Transformers with Spatial-Temporal Token Selection

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wang, Junke Yang, Xitong Li, Hengduo Liu, Li Wu, Zuxuan Jiang, Yu-Gang Shanghai Key Lab of Intell. Info. Processing School of CS Fudan University China Shanghai Collaborative Innovation Center on Intelligent Visual Computing China University of Maryland United States BirenTech Research

Video transformers have achieved impressive results on major video recognition benchmarks, which however suffer from high computational cost. In this paper, we present STTS, a token selection framework that dynamically selects a few informative tokens in both temporal and spatial dimensions conditioned on input video samples. Specifically, we formulate token selection as a ranking problem, which estimates the importance of each token through a lightweight scorer network and only those with top scores will be used for downstream evaluation. In the temporal dimension, we keep the frames that are most relevant to the action categories, while in the spatial dimension, we identify the most discriminative region in feature maps without affecting the spatial context used in a hierarchical way in most video transformers. Since the decision of token selection is non-differentiable, we employ a perturbed-maximum based differentiable Top-K operator for end-to-end training. We mainly conduct extensive experiments on Kinetics-400 with a recently introduced video transformer backbone, MViT. Our framework achieves similar results while requiring 20% less computation. We also demonstrate our approach is generic for different transformer architectures and video datasets. Code is available at https://***/wangjk666/STTS. Copyright © 2021, The Authors. All rights reserved.

关键词： Personnel training

Boosting the Transferability of Adversarial Attacks with Global Momentum Initialization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Jiafeng Chen, Zhaoyu Jiang, Kaixun Yang, Dingkang Hong, Lingyi Guo, Pinxue Guo, Haijing Zhang, Wenqiang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University Shanghai200433 China Engineering Research Center of Robotics Ministry of Education Academy for Engineering & Technology Fudan University Shanghai200433 China

Deep Neural Networks (DNNs) are vulnerable to adversarial examples, which are crafted by adding human-imperceptible perturbations to the benign inputs. Simultaneously, adversarial examples exhibit transferability across models, enabling practical black-box attacks. However, existing methods are still incapable of achieving the desired transfer attack performance. In this work, focusing on gradient optimization and consistency, we analyse the gradient elimination phenomenon as well as the local momentum optimum dilemma. To tackle these challenges, we introduce Global Momentum Initialization (GI), providing global momentum knowledge to mitigate gradient elimination. Specifically, we perform gradient pre-convergence before the attack and a global search during this stage. GI seamlessly integrates with existing transfer methods, significantly improving the success rate of transfer attacks by an average of 6.4% under various advanced defense mechanisms compared to the state-of-the-art method. Ultimately, GI demonstrates strong transferability in both image and video attack domains. Particularly, when attacking advanced defense methods in the image domain, it achieves an average attack success rate of 95.4%. The code is available at https://***/Omenzychen/Global-Momentum-Initialization. Copyright © 2022, The Authors. All rights reserved.

关键词： Deep neural networks

Symbiotic evolution of photonics and artificial intelligence: a comprehensive review

学校读者我要写书评

暂无评论

advanced Photonics 2025年第2期 8-85页

作者： Fu Feng Dewang Huo Ziyang Zhang Yijie Lou Shengyao Wang Zhijuan Gu Dong-Sheng Liu Xinhui Duan Daqian Wang Xiaowei Liu Ji Qi Shaoliang Yu Qingyang Du Guangyong Chen Cuicui Lu Yu Yu Xifeng Ren Xiaocong Yuan Zhejiang Lab Research Center for Frontier Fundamental Studies Westlake Institute for Optoelectronics Zhejiang Key Laboratory of 3D Micro/Nano Fabrication and Characterization Beijing Institute of Technology School of Physics Center for Interdisciplinary Science of Optical Quantum and NEMS Integration Key Laboratory of Advanced Optoelectronic Quantum Architecture and Measurements of Ministry of Education Beijing Key Laboratory of Nanophotonics and Ultrafine Optoelectronic Systems Huazhong University of Science and Technology Wuhan National Laboratory for Optoelectronics University of Science and Technology of China CAS Key Laboratory of Quantum Information University of Science and Technology of China CAS Center for Excellence in Quantum Information and Quantum Physics Zhejiang Lab Research Center for Life Sciences Computing

The rapid advancement of artificial intelligence(AI) has significantly impacted photonics, creating a symbiotic relationship that accelerates the development and applications of both fields. From the perspective of AI aiding photonics, deep-learning methods and various intelligent algorithms have been developed for designing complex photonic structures, where traditional design approaches fall short. AI's capability to process and analyze large data sets has enabled the discovery of novel materials, such as for photovoltaics,leading to enhanced light absorption and efficiency. AI is also significantly transforming the field of optical imaging with improved performance. In addition, AI-driven techniques have revolutionized optical communication systems by optimizing signal processing and enhancing the bandwidth and reliability of data transmission. Conversely, the contribution of photonics to AI is equally profound. Photonic technologies offer unparalleled advantages in the development of AI hardware, providing solutions to overcome the bottlenecks of electronic systems. The implementation of photonic neural networks, leveraging the high speed and parallelism of optical computing, demonstrates significant improvements in the processing speed and energy efficiency of AI computations. Furthermore, advancements in optical sensors and imaging technologies not only enrich AI applications with high-quality data but also expand the capabilities of AI in fields such as autonomous vehicles and medical imaging. We provide comprehensive knowledge and a detailed analysis of the current state of the art, addressing both challenges and opportunities at the intersection of AI and photonics. The multifaceted interactions between AI and photonics will be explored, illustrating how AI has become an indispensable tool in the development of photonics and how photonics, in turn,facilitates advancements in AI. Through a collection of case studies and examples, we underscore the potential

关键词：

KADEL: Knowledge-Aware Denoising Learning for Commit Message Generation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Tao, Wei Zhou, Yucheng Wang, Yanlin Zhang, Hongyu Wang, Haofen Zhang, Wenqiang Shanghai Engineering Research Center of AI and Robotics Academy for Engineering and Technology Fudan University Shanghai China State Key Laboratory of Internet of Things for Smart City Department of Computer and Information Science University of Macau China School of Software Engineering Sun Yat-sen University Guangdong Zhuhai519082 China School of Big Data and Software Engineering Chongqing University Chongqing China College of Design and Innovation Tongji University Shanghai China Engineering Research Center of AI and Robotics Ministry of Education Academy for Engineering and Technology Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University 220 Handan Road Shanghai200433 China

Commit messages are natural language descriptions of code changes, which are important for software evolution such as code understanding and maintenance. However, previous methods are trained on the entire dataset without considering the fact that a portion of commit messages adhere to good practice (i.e., good-practice commits), while the rest do not. On the basis of our empirical study, we discover that training on good-practice commits significantly contributes to the commit message generation. Motivated by this finding, we propose a novel knowledge-aware denoising learning method called KADEL. Considering that good-practice commits constitute only a small proportion of the dataset, we align the remaining training samples with these good-practice commits. To achieve this, we propose a model that learns the commit knowledge by training on good-practice commits. This knowledge model enables supplementing more information for training samples that do not conform to good practice. However, since the supplementary information may contain noise or prediction errors, we propose a dynamic denoising training method. This method composes a distribution-aware confidence function and a dynamic distribution list, which enhances the effectiveness of the training process. Experimental results on the whole MCMD dataset demonstrate that our method overall achieves state-of-the-art performance compared with previous methods. © 2024, CC BY.

关键词： Sampling

Automated Detection of Epileptic Spikes and Seizures Incorporating a Novel Spatial Clustering Prior

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Dong, Hanyang Sheng, Shurong Wang, Xiongfei Gao, Jia-Hong Sun, Yi Yang, Wanli Xiao, Kuntao Teng, Pengfei Luan, Guoming Lv, Zhao School of Artificial Intelligence Anhui University Hefei230601 China Anhui Province Key Laboratory of Biomedical Imaging and Intelligent Processing Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei230088 China Department of Neurosurgery Beijing Key Laboratory of Epilepsy Sanbo Brain Hospital Capital Medical University Beijing China McGovern Institute for Brain Research Peking University Beijing China Center for MRI Research Academy for Advanced Interdisciplinary Studies Peking University Beijing China Beijing City Key Lab for Medical Physics and Engineering Institute of Heavy Ion Physics School of Physics Peking University Beijing China Anhui Province Key Laboratory of Multimodal Cognitive Computation School of Computer Science and Technology Anhui University Hefei230601 China

A Magnetoencephalography (MEG) time-series recording consists of multi-channel signals collected by superconducting sensors, with each signal’s intensity reflecting magnetic field changes over time at the sensor location. Automating epileptic MEG spike detection significantly reduces manual assessment time and effort, yielding substantial clinical benefits. Existing research addresses MEG spike detection by encoding neural network inputs with signals from all channel within a time segment, followed by classification. However, these methods overlook simultaneous spiking occurred from nearby sensors. We introduce a simple yet effective paradigm that first clusters MEG channels based on their sensor’s spatial position. Next, a novel convolutional input module is designed to integrate the spatial clustering and temporal changes of the signals. This module is fed into a custom MEEG-ResNet3D developed by the authors, which learns to extract relevant features and classify the input as a spike clip or not. Our method achieves an F1 score of 94.73% on a large real-world MEG dataset Sanbo-CMR collected from two centers, outperforming state-of-the-art approaches by 1.85%. Moreover, it demonstrates efficacy and stability in the Electroencephalographic (EEG) seizure detection task, yielding an improved weighted F1 score of 1.4% compared to current state-of-the-art techniques evaluated on TUSZ, whch is the largest EEG seizure dataset. Copyright © 2025, The Authors. All rights reserved.

关键词： Magnetoencephalography

Characterizing Multi-domain False News and Underlying User Effects on Chinese Weibo

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Sheng, Qiang Cao, Juan Bernard, H. Russell Shu, Kai Li, Jintao Liu, Huan Key Lab of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China University of Chinese Academy of Sciences Beijing China Institute for Social Science Research Arizona State University TempeAZ United States Department of Computer Science Illinois Institute of Technology ChicagoIL United States Computer Science and Engineering Arizona State University TempeAZ United States

False news that spreads on social media has proliferated over the past years and has led to multi-aspect threats in the real world. While there are studies of false news on specific domains (like politics or health care), little work is found comparing false news across domains. In this article, we investigate false news across nine domains on Weibo, the largest Twitter-like social media platform in China, from 2009 to 2019. The newly collected data comprise 44,728 posts in the nine domains, published by 40,215 users, and reposted over 3.4 million times. Based on the distributions and spreads of the multi-domain dataset, we observe that false news in domains that are close to daily life like health and medicine generated more posts but diffused less effectively than those in other domains like politics, and that political false news had the most effective capacity for diffusion. The widely diffused false news posts on Weibo were associated strongly with certain types of users - by gender, age, etc. Further, these posts provoked strong emotions in the reposts and diffused further with the active engagement of false-news starters. Our findings have the potential to help design false news detection systems in suspicious news discovery, veracity prediction, and display and explanation. The comparison of the findings on Weibo with those of existing work demonstrates nuanced patterns, suggesting the need for more research on data from diverse platforms, countries, or languages to tackle the global issue of false news. The code and new anonymized dataset are available at https://***/ICTMCG/Characterizing-Weibo-Multi-Domain-False-News. © 2022, CC BY-NC-ND.

关键词： Social networking (online)

The Minority Matters: A Diversity-Promoting Collaborative Metric Learning Algorithm

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Bao, Shilong Xu, Qianqian Yang, Zhiyong He, Yuan Cao, Xiaochun Huang, Qingming State Key Laboratory of Information Security Institute of Information Engineering CAS China School of Cyber Security University of Chinese Academy of Sciences China Key Lab. of Intelligent Information Processing Institute of Computing Technology CAS China School of Computer Science and Tech. University of Chinese Academy of Sciences China Alibaba Group China School of Cyber Science and Technology Shenzhen Campus Sun Yat-sen University China Key Laboratory of Big Data Mining and Knowledge Management CAS China Peng Cheng Laboratory China

Collaborative Metric Learning (CML) has recently emerged as a popular method in recommendation systems (RS), closing the gap between metric learning and Collaborative Filtering. Following the convention of RS, existing methods exploit unique user representation in their model design. This paper focuses on a challenging scenario where a user has multiple categories of interests. Under this setting, we argue that the unique user representation might induce preference bias, especially when the item category distribution is imbalanced. To address this issue, we propose a novel method called Diversity-Promoting Collaborative Metric Learning (DPCML), with the hope of considering the commonly ignored minority interest of the user. The key idea behind DPCML is to include a multiple set of representations for each user in the system. Based on this embedding paradigm, user preference toward an item is aggregated from different embeddings by taking the minimum item-user distance among the user embedding set. Furthermore, we observe that the diversity of the embeddings for the same user also plays an essential role in the model. To this end, we propose a diversity control regularization term to accommodate the multi-vector representation strategy better. Theoretically, we show that DPCML could generalize well to unseen test data by tackling the challenge of the annoying operation that comes from the minimum value. Experiments over a range of benchmark datasets speak to the efficacy of DPCML. © 2022, CC BY-NC-SA.

关键词： Embeddings

Utility-Oriented Underwater Image Quality Assessment Based on Transfer Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Chen, Weiling Lin, Rongfu Liao, Honggang Zhao, Tiesong Gu, Ke Le Callet, Patrick Fujian Key Lab for Intelligent Processing and Wireless Transmission of Media Information Fuzhou University Fuzhou350108 China Peng Cheng Laboratory Shenzhen518000 China The Faculty of Information Technology Beijing University of Technology Engineering Research Center of Intelligent Perception and Autonomous Control Ministry of Education Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing Artificial Intelligence Institute Beijing100124 China Équipe Image Perception et Interaction Laboratoire des Sciences du Numérique de Nantes Université de Nantes France

The widespread image applications have greatly promoted the vision-based tasks, in which the Image Quality Assessment (IQA) technique has become an increasingly significant issue. For user enjoyment in multimedia systems, the IQA exploits image fidelity and aesthetics to characterize user experience;while for other tasks such as popular object recognition, there exists a low correlation between utilities and perceptions. In such cases, the fidelity-based and aesthetics-based IQA methods cannot be directly applied. To address this issue, this paper proposes a utility-oriented IQA in object recognition. In particular, we initialize our research in the scenario of underwater fish detection, which is a critical task that has not yet been perfectly addressed. Based on this task, we build an Underwater Image Utility Database (UIUD) and a learning-based Underwater Image Utility Measure (UIUM). Inspired by the top-down design of fidelity-based IQA, we exploit the deep models of object recognition and transfer their features to our UIUM. Experiments validate that the proposed transfer-learning-based UIUM achieves promising performance in the recognition task. We envision our research provides insights to bridge the researches of IQA and computer vision. Copyright © 2022, The Authors. All rights reserved.

关键词： Multimedia systems