检索结果-内蒙古大学图书馆

Multi-label Classification with High-rank and High-order Label Correlations

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Si, Chongjie Jia, Yuheng Wang, Ran Zhang, Min-Ling Feng, Yanghe Qu, Chongxiao The Chien-Shiung Wu College Southeast University Nanjing210096 China The MoE Key Lab of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai200240 China The School of Computer Science and Engineering Southeast University Nanjing210096 China Ministry of Education China School of Computing & Information Sciences Caritas Institute of Higher Education Hong Kong The Key Laboratory of Computer Network and Information Integration Southeast University Ministry of Education China The School of Mathematical Science Shenzhen University Shenzhen518060 China Shenzhen Key Laboratory of Advanced Machine Learning and Applications Shenzhen University Shenzhen518060 China The College of Systems Engineering National University of Defense Technology China The 52nd Research Institute of China Electronics Technology Group China

Exploiting label correlations is important to multi-label classification. Previous methods capture the high-order label correlations mainly by transforming the label matrix to a latent label space with low-rank matrix factorization. However, the label matrix is generally a full-rank or approximate full-rank matrix, making the low-rank factorization inappropriate. Besides, in the latent space, the label correlations will become implicit. To this end, we propose a simple yet effective method to depict the high-order label correlations explicitly, and at the same time maintain the high-rank of the label matrix. Moreover, we estimate the label correlations and infer model parameters simultaneously via the local geometric structure of the input to achieve mutual enhancement. Comparative studies over twelve benchmark data sets validate the effectiveness of the proposed algorithm in multi-label classification. The exploited high-order label correlations are consistent with common sense empirically. Our code is publicly available at https://***/Chongjie-Si/HOMI. Copyright © 2022, The Authors. All rights reserved.

关键词： Matrix algebra

APPFLx: Providing Privacy-Preserving Cross-Silo Federated Learning as a Service

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Li, Zilinghan He, Shilan Chaturvedi, Pranshu Hoang, Trung-Hieu Ryu, Minseok Huerta, E.A. Kindratenko, Volodymyr Fuhrman, Jordan Giger, Maryellen Chard, Ryan Kim, Kibaek Madduri, Ravi Data Science and Learning Division Argonne National Laboratory LemontIL60439 United States Department of Computer Science University of Illinois at Urbana-Champaign UrbanaIL61801 United States National Center for Supercomputing Applications University of Illinois at Urbana-Champaign UrbanaIL61801 United States Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign UrbanaIL61801 United States School of Computing and Augmented Intelligence Arizona State University TempeAZ85281 United States Department of Physics University of Illinois at Urbana-Champaign UrbanaIL61801 United States Department of Computer Science The University of Chicago ChicagoIL60637 United States University of Chicago Consortium for Advanced Science and Engineering ChicagoIL60637 United States Mathematics and Computer Science Division Argonne National Laboratory LemontIL60439 United States Department of Radiology University of Chicago Hyde ParkIL60616 United States

Cross-silo privacy-preserving federated learning (PPFL) is a powerful tool to collaboratively train robust and generalized machine learning (ML) models without sharing sensitive (e.g., healthcare of financial) local data. To ease and accelerate the adoption of PPFL, we introduce APPFLx, a ready-to-use platform that provides privacy-preserving cross-silo federated learning as a service. APPFLx employs Globus authentication to allow users to easily and securely invite trustworthy collaborators for PPFL, implements several synchronous and asynchronous FL algorithms, streamlines the FL experiment launch process, and enables tracking and visualizing the life cycle of FL experiments, allowing domain experts and ML practitioners to easily orchestrate and evaluate cross-silo FL under one platform. APPFLx is available online at https://*** Index Terms—Federated learning, privacy preserving, federation as a service, AI for science, science as a service, HPC, IAM Copyright © 2023, The Authors. All rights reserved.

关键词： Life cycle

Unveiling Hidden Vulnerabilities in Digital Human Generation via Adversarial Attacks

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Li, Zhiying Jin, Yeying Shen, Fan Liu, Zhi Chen, Weibin Zhang, Pengju Zhang, Xiaomei Chen, Boyu Shen, Michael Wu, Kejian Fan, Zhaoxin Dong, Jin College of Cyber Security Jinan University China Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing School of Artificial Intelligence Beihang University China Hangzhou International Innovation Institute Beihang University China Department of Electrical and Computer Engineering National University of Singapore Singapore Department of Electrical and Computer Science University of Pittsburgh United States State Key Laboratory of Multimodal Artificial Intelligence Institute of Automation Chinese Academy of Science China Institute of Automation Chinese Academy of Science China University College London United Kingdom Mingdu Tech China Xreal Beijing Academy of Blockchain and Edge Computing China

Expressive human pose and shape estimation (EHPS) is crucial for digital human generation, especially in applications like live streaming. While existing research primarily focuses on reducing estimation errors, it largely neglects robustness and security aspects, leaving these systems vulnerable to adversarial attacks. To address this significant challenge, we propose the Tangible Attack (TBA), a novel framework designed to generate adversarial examples capable of effectively compromising any digital human generation model. Our approach introduces a Dual Heterogeneous Noise Generator (DHNG), which leverages Variational Autoencoders (VAE) and ControlNet to produce diverse, targeted noise tailored to the original image features. Additionally, we design a custom adversarial loss function to optimize the noise, ensuring both high controllability and potent disruption. By iteratively refining the adversarial sample through multi-gradient signals from both the noise and the state-of-the-art EHPS model, TBA substantially improves the effectiveness of adversarial attacks. Extensive experiments demonstrate TBA’s superiority, achieving a remarkable 41.0% increase in estimation error, with an average improvement of approximately 17.0%. These findings expose significant security vulnerabilities in current EHPS models and highlight the need for stronger defenses in digital human generation systems. Copyright © 2025, The Authors. All rights reserved.

关键词： Man machine systems

Multi-stage Speaker Extraction with Utterance and Frame-Level Reference Signals

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Ge, Meng Xu, Chenglin Wang, Longbiao Chng, Eng Siong Dang, Jianwu Li, Haizhou Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China School of Computer Science and Engineering Nanyang Technological University Singapore Japan Advanced Institute of Science and Technology Ishikawa Japan Department of Electrical and Computer Engineering National University of Singapore Singapore Machine Listening Lab University of Bremen Germany

Speaker extraction requires a sample speech from the target speaker as the reference. However, enrolling a speaker with a long speech is not practical. We propose a speaker extraction technique, that performs in multiple stages to take full advantage of short reference speech sample. The extracted speech in early stages is used as the reference speech for late stages. For the first time, we use frame-level sequential speech embedding as the reference for target speaker. This is a departure from the traditional utterance-based speaker embedding reference. In addition, a signal fusion scheme is proposed to combine the decoded signals in multiple scales with automatically learned weights. Experiments on WSJ0-2mix and its noisy versions (WHAM! and WHAMR!) show that SpEx++ consistently outperforms other state-of-the-art baselines. Copyright © 2020, The Authors. All rights reserved.

关键词： Embeddings

$\mathsf{NCF}$NCF: A Neural Context Fusion Approach to Raw Mobility Annotation

学校读者我要写书评

暂无评论

IEEE Transactions on Mobile computing 2020年第1期21卷 226-238页

作者： Renjun Hu Jingbo Zhou Xinjiang Lu Hengshu Zhu Shuai Ma Hui Xiong SKLSDE Lab Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing China Business Intelligence Lab Baidu Research National Engineering Laboratory of Deep Learning Technology and Application Beijing China Talent Intelligence Center Baidu Inc. Beijing China Management Science and Information Systems Department Rutgers Business School Rutgers University Newark NJ USA

Understanding human mobility patterns at the point-of-interest (POI) scale plays an important role in enhancing business intelligence in mobile environments. While large efforts have been made in this direction, most studies simply utilize POI check-ins to mine the concerned mobility patterns, the effectiveness of which is usually hindered due to data sparsity. To obtain better POI-based human mobility for mining, in this paper, we strive to directly annotate the POIs associated with raw user-generated mobility records. We propose a neural context fusion approach which integrates various context factors in people's POI-visiting behaviors. Our approach evaluates the preference and transition factors via representation learning. Notably, we incorporate an attention mechanism to deal with the randomized transitions in raw mobility. The domain knowledge factors, i.e., distance, time and popularity, remain effective and our approach further includes them from a data-driven perspective. Factors are automatically fused with a feed-forward neural network. Furthermore, we exploit a multi-head architecture to enhance the model expressiveness. Using two real-life data sets, we conduct our experimental study and find that our approach consistently outperforms the state-of-the-art baselines by at least 32 percent in accuracy. Besides, we demonstrate the utility of the obtained POI-based human mobility with a POI recommendation example.

关键词： Annotations Trajectory Mobile computing Computer architecture Business intelligence Neural networks Semantics

Project Hephaistos - II. Dyson sphere candidates from Gaia DR3, 2MASS, and WISE

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Suazo, Matías Zackrisson, Erik Mahto, Priyatam K. Lundell, Fabian Nettelblad, Carl Korn, Andreas J. Wright, Jason T. Majumdar, Suman Observational Astrophysics Department of Physics and Astronomy Uppsala University Box 516 UppsalaSE-751 20 Sweden Department of Astronomy Astrophysics and Space Engineering Indian Institute of Technology Indore India Division of Scientific Computing Department of Information Technology Uppsala University UppsalaSE-751 05 Sweden Department of Astronomy & Astrophysics The Pennsylvania State University University ParkPA16802 United States Penn State Extraterrestrial Intelligence Center The Pennsylvania State University University ParkPA16802 United States Center for Exoplanets and Habitable Worlds The Pennsylvania State University University ParkPA16802 United States Department of Physics Blackett Laboratory Imperial College LondonSW7 2AZ United Kingdom Swedish Collegium for Advanced Study Linneanum Thunbergsvägen 2 UppsalaSE-752 38 Sweden

The search for extraterrestrial intelligence is currently being pursued using multiple techniques and in different wavelength bands. Dyson spheres, megastructures that could be constructed by advanced civilizations to harness the radiation energy of their host stars, represent a potential technosignature, that in principle may be hiding in public data already collected as part of large astronomical surveys. In this study, we present a comprehensive search for partial Dyson spheres by analyzing optical and infrared observations from Gaia, 2MASS, and WISE. We develop a pipeline that employs multiple filters to identify potential candidates and reject interlopers in a sample of five million objects, which incorporates a convolutional neural network to help identify confusion in WISE data. Finally, the pipeline identifies 7 candidates deserving of further analysis. All of these objects are M-dwarfs, for which astrophysical phenomena cannot easily account for the observed infrared excess emission. Copyright © 2024, The Authors. All rights reserved.

关键词： Pipelines

A Multi-Task Mean Teacher for Semi-Supervised Shadow Detection

学校读者我要写书评

暂无评论

A Multi-Task Mean Teacher for Semi-Supervised Shadow Detecti...

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Zhihao Chen Lei Zhu Liang Wan Song Wang Wei Feng Pheng-Ann Heng College of Intelligence and Computing Tianjin University Department of Computer Science and Engineering The Chinese University of Hong Kong Department of Computer Science and Engineering University of South Carolina Shenzhen Key Laboratory of Virtual Reality and Human Interaction Technology Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences China

ISBN: (数字)9781728171685

ISBN: (纸本)9781728171692

Existing shadow detection methods suffer from an intrinsic limitation in relying on limited labeled datasets, and they may produce poor results in some complicated situations. To boost the shadow detection performance, this paper presents a multi-task mean teacher model for semi-supervised shadow detection by leveraging unlabeled data and exploring the learning of multiple information of shadows simultaneously. To be specific, we first build a multi-task baseline model to simultaneously detect shadow regions, shadow edges, and shadow count by leveraging their complementary information and assign this baseline model to the student and teacher network. After that, we encourage the predictions of the three tasks from the student and teacher networks to be consistent for computing a consistency loss on unlabeled data, which is then added to the supervised loss on the labeled data from the predictions of the multi-task baseline model. Experimental results on three widely-used benchmark datasets show that our method consistently outperforms all the compared state-of- the-art methods, which verifies that the proposed network can effectively leverage additional unlabeled data to boost the shadow detection performance.

关键词： Task analysis Image edge detection Feature extraction Training Radio frequency Fuses Solid modeling

Free Viewpoint Video Synthesis Based on DIBR

学校读者我要写书评

暂无评论

Free Viewpoint Video Synthesis Based on DIBR

IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)

作者： Xuesong Gao Keqiu Li Weiqiang Chen Zhongyao Yang Wenguo Wei Yangang Cai College of Intelligence and Computing Tianjin University Tianjin China Tianjin Key Laboratory of Advanced Networking (TANK) Tianjin China State Key Laboratory of Digital Multimedia Technology Hisense Co. Ltd. Qingdao China School of Electronics Engineering and Computer Science Peking University Beijing China

ISBN: (数字)9781728142722

ISBN: (纸本)9781728142739

The Free Viewpoint Video (FVV) system is an advanced visual media mode, which could satisfy the audience's demand of watching from any viewpoints. It has aroused great interests in the field of multimedia and vision. This paper proposes a set of solutions, including depth estimation, virtual view interpolation and post processing, to synthesize and optimize the virtual view images. Specifically, deep learning based method is proposed to compute depth images, which is more efficient than traditional stereo matching algorithm. The virtual view can be synthesized with the help of DIBR algorithm in an arbitrary position between two reference cameras. To remove the noise caused by rendering, some post processing techniques, such as median filtering and hole filling, are employed. The extensive experimental results demonstrate that the quality of the synthesized free viewpoint video has been significantly improved using our method.

关键词： Estimation Image color analysis Cameras Three-dimensional displays Conferences Rendering (computer graphics) Streaming media

Efficient and Privacy-Preserving Speaker Verification Scheme for Home Automation Devices

学校读者我要写书评

暂无评论

Efficient and Privacy-Preserving Speaker Verification Scheme...

IEEE Conference on Multimedia Information Processing and Retrieval (MIPR)

作者： Xuesong Gao Keqiu Li Weiqiang Chen Weifeng Hu Ziling Zhang Qi Li College of Intelligence and Computing Tianjin University Tianjin China Tianjin Key Laboratory of Advanced Networking (TANK) Tianjin China State Key Laboratory of Digital Multimedia Technology Hisense Co. Ltd. Qingdao China School of Information Science and Engineering Shandong University Qingdao China School of Cyber Engineering Xidian University Xi'an China

ISBN: (数字)9781728142722

ISBN: (纸本)9781728142739

Speaker recognition technology has been widely used in home automation recently, which can provide identification for the user efficiently. However, speaker recognition technology brings security risks as well as convenience, especially the privacy disclosure of the user's voiceprint templates. This paper proposes an efficient and privacy-preserving speaker verification scheme for home automation devices, named SEASON. We modify the high-accuracy deepspeaker framework in terms of privacy, based on residual network. With SEASON, users can access the home automation devices securely. Detailed security analysis shows that SEASON can actually resist various security threats. We implement SEASON with real datasets and 48 real records from four testers. Experiments reveal that SEASON can achieve efficient and accurate verification.

关键词： Spectrogram Authentication Home automation Privacy IEEE transactions Speaker recognition