检索结果-内蒙古大学图书馆

Audio-Visual Generalized Zero-Shot Learning Based on Variational Information Bottleneck

学校读者我要写书评

暂无评论

Audio-Visual Generalized Zero-Shot Learning Based on Variati...

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yapeng Li Yong Luo Bo Du School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence and Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan China

Audio-visual generalized zero-shot learning (GZSL) aims to train a model on seen classes for classifying data samples from both seen classes and unseen classes. Due to the absence of unseen training samples, the model tends to misclassify unseen class samples into seen classes. To mitigate this problem, in this paper, we propose a method based on variational information bottleneck for audio-visual GZSL. Specifically, we model the joint representations as a product-of-experts over marginal representations to integrate the information of audio and visual. Besides, we introduce variational information bottleneck to the learning of audio-visual joint representations and marginal representations of audio, visual, and text label modalities. This helps our model reduce the negative impact of information that cannot be generalized to unseen classes. Experimental results conducted on the UCF-GZSL, VGGSound-GZSL, and ActivityNet-GZSL benchmarks demonstrate the effectiveness and superiority of the proposed model for audio-visual GZSL.

关键词：

FriendsQA: A New Large-Scale Deep Video Understanding Dataset with Fine-grained Topic Categorization for Story Videos

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Zhengqian Li, Ruizhe Xu, Zijun Wang, Zhongyuan Xiao, Chunxia Liang, Chao School of Computer Science Wuhan University China School of Cyber Science and Engineering Wuhan University China National Engineering Research Center for Multimedia Software Wuhan University China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China

Video question answering (VideoQA) aims to answer natural language questions according to the given videos. Although existing models perform well in the factoid VideoQA task, they still face challenges in deep video understanding (DVU) task, which focuses on story videos. Compared to factoid videos, the most significant feature of story videos is storylines, which are composed of complex interactions and long-range evolvement of core story topics including characters, actions and locations. Understanding these topics requires models to possess DVU capability. However, existing DVU datasets rarely organize questions according to these story topics, making them difficult to comprehensively assess VideoQA models' DVU capability of complex storylines. Additionally, the question quantity and video length of these dataset are limited by high labor costs of handcrafted dataset building method. In this paper, we devise a large language model based multi-agent collaboration framework, StoryMind, to automatically generate a new large-scale DVU dataset. The dataset, FriendsQA, derived from the renowned sitcom Friends with an average episode length of 1,358 seconds, contains 44.6K questions evenly distributed across 14 fine-grained topics. Finally, We conduct comprehensive experiments on 10 state-of-the-art VideoQA models using the FriendsQA dataset. © 2024, CC BY.

关键词： Wages

A CPK-Based Identity Authentication Scheme for IoT

学校读者我要写书评

暂无评论

computer Systems Science & engineering 2022年第3期40卷 1217-1231页

作者： Mingming Zhang Jiaming Mao Yuanyuan Ma Liangjie Xu Chuanjun Wang Ran Zhao Zhihao Li Lu Chen Wenbing Zhao State Grid Jiangsu Electric Power Co. Ltd.Information&Telecommunication BranchNanjing210008China State Grid Key Laboratory of Information&Network Security Institute of Information and CommunicationGlobal Energy Interconnection Research InstituteNanjing210003China Anhui Jiyuan Software Co.Ltd SGITGHefei230088China Engineering Research Center of Post Big Data Technology and Application of Jiangsu Province Research and Development Center of Post Industry Technology of the State Posts Bureau(Internet of Things Technology)Engineering Research Center of Broadband Wireless Communication Technology of the Ministry of EducationNanjing University of Posts and TelecommunicationsNanjing210003China Department of Electrical Engineering and Computer Science Cleveland State UniversityCleveland44115USA

As the power Internet of Things(IoT)enters the security construction stage,the massive use of perception layer devices urgently requires an identity authentication scheme that considers both security and *** existing public key infrastructure(PKI)-based security authentication scheme is currently difficult to apply in many terminals in *** key distribution and management costs are high,which hinders the development of power IoT security *** Public key(CPK)technology uses a small number of seeds to generate unlimited public *** is very suitable for identity authentication in the power Internet of *** this paper,we propose a novel identity authentication scheme for power *** scheme combines the physical unclonable function(PUF)with improved CPK technology to achieve mutual identity authentication between power IoT terminals and *** proposed scheme does not require third-party authentication and improves the security of identity authentication for power ***,the scheme reduces the resource consumption of power IoT *** improved CPK algorithm solves the key collision problem,and the third party only needs to save the private key and the public key *** results show that the amount of storage resources occupied in our scheme is *** proposed scheme is more suitable for the power IoT.

关键词： Power Internet of Things terminal equipment identity authentication CPK PUF

TPE-H2MWD:an exact thumbnail preserving encryption scheme with hidden Markov model and weighted diffusion

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic engineering 2023年第8期24卷 1169-1180页

作者： Xiuli CHAI Xiuhui CHEN Yakun MA Fang ZUO Zhihua GAN Yushu ZHANG School of Artificial Intelligence Henan Engineering Research Center for Industrial Internet of ThingsHenan UniversityZhengzhou450046China Henan Key Laboratory of Cyberspace Situation Awareness Zhengzhou450001China School of Software Intelligent Data Processing Engineering Research Center of Henan ProvinceInstitute of Intelligent Network SystemHenan UniversityKaifeng475004China College of Computer Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing211106China

With the substantial increase in image transmission,the demand for image security is ***-like images can be obtained by conventional encryption schemes,and although the security of the images can be guaranteed,the noise-like images cannot be directly previewed and *** on the rank-then-encipher method,some researchers have designed a three-pixel exact thumbnail preserving encryption(TPE2)scheme,which can be applied to balance the security and availability of images,but this scheme has low encryption *** this paper,we introduce an efficient exact thumbnail preserving encryption ***,blocking and bit-plane decomposition operations are performed on the plaintext *** zigzag scrambling model is used to change the bit positions in the lower four bit ***,an operation is devised to permute the higher four bit planes,which is an extended application of the hidden Markov ***,according to the difference in bit weights in each bit plane,a bit-level weighted diffusion rule is established to generate an encrypted image and still maintain the same sum of pixels within the *** results show that the proposed scheme improves the encryption efficiency and can guarantee the availability of images while protecting their privacy.

关键词： Hidden Markov model Weighted diffusion Balance between usability and privacy Image encryption

Local Features Meet Stochastic Anonymization: Revolutionizing Privacy-Preserving Face Recognition for Black-Box Models

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Yuanwei Jia, Chengyu Xiao, Ruqi Jia, Xuemei Wei, Hui Jiang, Kui Wang, Zheng National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Wuhan University China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China School of Computer Science and Technology Harbin Institute of Technology Harbin China

The task of privacy-preserving face recognition (PPFR) currently faces two major unsolved challenges: (1) existing methods are typically effective only on specific face recognition models and struggle to generalize to black-box face recognition models;(2) current methods employ data-driven reversible representation encoding for privacy protection, making them susceptible to adversarial learning and reconstruction of the original image. We observe that face recognition models primarily rely on local features (e.g., face contour, skin texture, and so on) for identification. Thus, by disrupting global features while enhancing local features, we achieve effective recognition even in black-box environments. Additionally, to prevent adversarial models from learning and reversing the anonymization process, we adopt an adversarial learning-based approach with irreversible stochastic injection to ensure the stochastic nature of the anonymization. Experimental results demonstrate that our method achieves an average recognition accuracy of 94.21% on black-box models, outperforming existing methods in both privacy protection and anti-reconstruction capabilities. Copyright © 2024, The Authors. All rights reserved.

关键词： Differential privacy

Verify All Traffic: Towards Zero-Trust In-network Intrusion Detection against Multipath Routing

学校读者我要写书评

暂无评论

IEEE Journal on Selected Areas in Communications 2025年第6期43卷 2155-2171页

作者： Zhao, Ziming Li, Zhaoxuan Xie, Xiaofei Liu, Zhipeng Li, Tingting Yu, Jiongchi Zhang, Fan Chen, Binbin School of Software Technology Zhejiang University Ningbo315100 China College of Computer Science and Technology Zhejiang University Hangzhou310027 China ZJU-Hangzhou Global Scientific and Technological Innovation Center 311200 China Key Laboratory of Blockchain and Cyberspace Governance of Zhejiang Province 310027 China Jiaxing Research Institute Zhejiang University 314000 China Zhengzhou Xinda Institute of Advanced Technology Zhengzhou450001 China Singapore State Key Laboratory of Information Security Institute of Information Engineering Chinese Academy of Sciences Beijing100093 China School of Cyber Security UCAS Beijing100049 China School of Computing and Information Systems Singapore Management University Singapore188065 Singapore Advanced Digital Sciences Center Singapore138632 Singapore Singapore University of Technology and Design SingaporeB96049 Singapore

With the popularity of encryption protocols, machine learning (ML)-based traffic analysis technologies have attracted widespread attention. To adapt to modern high-speed bandwidth, recent research is dedicated to advancing zero-trust intrusion detection by offloading feature extraction and model inference into the network dataplane. Especially, with the rise of programmable switches, achieving line-speed ML inference becomes promising. However, existing research only considers a single switch node as a relay to conduct evaluation. This is far from real-world deployments involving multiple switches (given that zero-trust security assumes that threats can originate from anywhere, including within the network), particularly the multi-path routing phenomenon that exists in practice. In this paper, we reveal practical challenges in the context of enabling line-speed model inference in the network dataplane. Furthermore, we propose FCPlane, the forwarding and computing integrated dataplane for zero-trust intrusion detection that aims to enable efficient load balancing while providing reliable traffic analysis results, even against multipath routing. The core idea is to reconcile forwarding and computation to the flowlet level, for which a tailor-made Markov chain model is designed. Based on two public traffic datasets, we evaluate seven state-of-the-art in-network traffic analysis models deployed in four types of topologies (three with multipath routing and one without) to explore performance impact and demonstrate the effectiveness of our proposal. © 1983-2012 IEEE.

关键词： network intrusion

All Information is Necessary: Integrating Speech Positive and Negative Information by Contrastive Learning for Speech Enhancement

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Xu, Xinmeng Tu, Weiping Han, Chang Yang, Yuhong National Engineering Research Center for Multimedia Software School of Computer Science Wuhan University China Hubei Luojia Laboratory China Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China

Monaural speech enhancement (SE) is an ill-posed problem due to the irreversible degradation process. Recent methods to achieve SE tasks rely solely on positive information, e.g., ground-truth speech and speech-relevant features. Different from the above, we observe that the negative information, such as original speech mixture and speech-irrelevant features, are valuable to guide the SE model training procedure. In this study, we propose a SE model that integrates both speech positive and negative information for improving SE performance by adopting contrastive learning, in which two innovations have consisted. (1) We design a collaboration module (CM), which contains two parts, contrastive attention for separating relevant and irrelevant features via contrastive learning and interactive attention for establishing the correlation between both speech features in a learnable and self-adaptive manner. (2) We propose a contrastive regularization (CR) built upon contrastive learning to ensure that the estimated speech is pulled closer to the clean speech and pushed far away from the noisy speech in the representation space by integrating self-supervised models. We term the proposed SE network with CM and CR as CMCR-Net. Experimental results demonstrate that our CMCR-Net achieves comparable and superior performance to recent approaches. © 2023, CC BY.

关键词： Speech enhancement

Learning from Imperfect Data: Towards Efficient Knowledge Distillation of Autoregressive Language Models for Text-to-SQL

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhong, Qihuang Chen, Kunfeng Ding, Liang Liu, Juhua Du, Bo Tao, Dacheng School of Computer Science National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China The University of Sydney Australia Nanyang Technological University Singapore

Large Language Models (LLMs) have shown promising performance in text-to-SQL, which involves translating natural language questions into SQL queries. However, current text-to-SQL LLMs are computationally expensive and challenging to deploy in real-world applications, highlighting the importance of compressing them. To achieve this goal, knowledge distillation (KD) is a common approach, which aims to distill the larger teacher model into a smaller student model. While numerous KD methods for autoregressive LLMs have emerged recently, it is still under-explored whether they work well in complex text-to-SQL scenarios. To this end, we conduct a series of analyses and reveal that these KD methods generally fall short in balancing performance and efficiency. In response to this problem, we propose to improve the KD with Imperfect Data, namely KID, which effectively boosts the performance without introducing much training budget. The core of KID is to efficiently mitigate the training-inference mismatch by simulating the cascading effect1 of inference in the imperfect training data. Extensive experiments on 5 text-to-SQL benchmarks show that, KID can not only achieve consistent and significant performance gains (up to +5.83% average score) across all model types and sizes, but also effectively improve the training efficiency. Copyright © 2024, The Authors. All rights reserved.

关键词： Structured Query Language

Dynamic personalized federated learning with adaptive differential privacy 23

学校读者我要写书评

暂无评论

Dynamic personalized federated learning with adaptive differ...

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Xiyuan Yang Wenke Huang Mang Ye National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence Hubei Key Laboratory of Multimedia and Network Communication Engineering School of Computer Science Hubei Luojia Laboratory Wuhan University Wuhan China

Personalized federated learning with differential privacy has been considered a feasible solution to address non-IID distribution of data and privacy leakage risks. However, current personalized federated learning methods suffer from inflexible personalization and convergence difficulties due to two main factors: 1) Firstly, we observe that the prevailing personalization methods mainly achieve this by personalizing a fixed portion of the model, which lacks flexibility. 2) Moreover, we further demonstrate that the default gradient calculation is sensitive to the widely-used clipping operations in differential privacy, resulting in difficulties in convergence. Considering that Fisher information values can serve as an effective measure for estimating the information content of parameters by reflecting the model sensitivity to parameters, we aim to leverage this property to address the aforementioned challenges. In this paper, we propose a novel federated learning method with Dynamic Fisher Personalization and Adaptive Constraint (FedDPA) to handle these challenges. Firstly, by using layer-wise Fisher information to measure the information content of local parameters, we retain local parameters with high Fisher values during the personalization process, which are considered informative, simultaneously prevent these parameters from noise perturbation. Secondly, we introduce an adaptive approach by applying differential constraint strategies to personalized parameters and shared parameters identified in the previous for better convergence. Our method boosts performance through flexible personalization while mitigating the slow convergence caused by clipping operations. Experimental results on CIFAR-10, FEMNIST and SVHN dataset demonstrate the effectiveness of our approach in achieving better performance and robustness against clipping, under personalized federated learning with differential privacy.

关键词：