检索结果-内蒙古大学图书馆

Vertical Federated Learning: Challenges, Methodologies and Experiments

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wei, Kang Li, Jun Ma, Chuan Ding, Ming Wei, Sha Wu, Fan Chen, Guihai Ranbaduge, Thilina School of Electrical and Optical Engineering Nanjing University of Science and Technology Nanjing210094 China Ministry of Education China Data61 CSIRO SydneyNSW2015 Australia Beijing100191 China The Shanghai Key Laboratory of Scalable Computing and Systems Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai200240 China The State Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China

Recently, federated learning (FL) has emerged as a promising distributed machine learning (ML) technology, owing to the advancing computational and sensing capacities of end-user devices, as well as the increasing concerns on users' privacy. As a special architecture in FL, vertical FL (VFL) is capable of constructing a hyper ML model by embracing sub-models from different types of clients. These sub-models are trained locally by vertically partitioned data with distinct attributes. Therefore, the design of VFL is fundamentally different from that of conventional FL, raising new and unique research issues. In this paper, we aim to discuss key challenges in VFL with effective solutions, and conduct experiments on real-life datasets to shed light on these issues. Specifically, we first propose a general framework on VFL, and highlight the key differences between VFL and conventional FL. Then, we discuss research challenges rooted in VFL systems from four aspects, i.e., security and privacy risks, expensive computation and communication costs, possible structural damage caused by model splitting, and system heterogeneity. Afterwards, we develop solutions to addressing the aforementioned challenges, and conduct extensive experiments to showcase the effectiveness of the proposed solutions. Copyright © 2022, The Authors. All rights reserved.

关键词： Federated learning

ExchNet: A Unified Hashing Network for Large-Scale Fine-Grained Image Retrieval 1

学校读者我要写书评

暂无评论

16th European Conference on computer Vision, ECCV 2020

作者： Cui, Quan Jiang, Qing-Yuan Wei, Xiu-Shen Li, Wu-Jun Yoshie, Osamu Graduate School of IPS Waseda University Fukuoka Japan National Key Laboratory for Novel Software Technology Department of Computer Science and Technology Nanjing University Nanjing China Megvii Research Nanjing Megvii Technology Nanjing China

ISBN: (数字)9783030585808

ISBN: (纸本)9783030585792

Retrieving content relevant images from a large-scale fine-grained dataset could suffer from intolerably slow query speed and highly redundant storage cost, due to high-dimensional real-valued embeddings which aim to distinguish subtle visual differences of fine-grained objects. In this paper, we study the novel fine-grained hashing topic to generate compact binary codes for fine-grained images, leveraging the search and storage efficiency of hash learning to alleviate the aforementioned problems. Specifically, we propose a unified end-to-end trainable network, termed as ExchNet. Based on attention mechanisms and proposed attention constraints, ExchNet can firstly obtain both local and global features to represent object parts and the whole fine-grained objects, respectively. Furthermore, to ensure the discriminative ability and semantic meaning’s consistency of these part-level features across images, we design a local feature alignment approach by performing a feature exchanging operation. Later, an alternating learning algorithm is employed to optimize the whole ExchNet and then generate the final binary hash codes. Validated by extensive experiments, our ExchNet consistently outperforms state-of-the-art generic hashing methods on five fine-grained datasets. Moreover, compared with other approximate nearest neighbor methods, ExchNet achieves the best speed-up and storage reduction, revealing its efficiency and practicality. © 2020, Springer Nature Switzerland AG.

关键词： Nearest neighbor search

Deep Forest with Hashing Screening and Window Screening

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ma, Pengfei Wu, Youxi Li, Yan Guo, Lei Jiang, He Zhu, Xingquan Wu, Xindong School of Artificial Intelligence Hebei University of Technology Tianjing300401 China Hebei Key Laboratory of Big Data Computing Tianjing300401 China School of Economics and Management Hebei University of Technology Tianjing300401 China State Key Laboratory of Reliability and Intelligence of Electrical Equipment Hebei University of Technology Tianjing300401 China School of Software Dalian University of Technology Dalian116023 China Department of Computer & Electrical Engineering and Computer Science Florida Atlantic University FL33431 United States Hefei University of Technology Hefei230009 China Mininglamp Academy of Sciences Mininglamp Technology Beijing100084 China

As a novel deep learning model, gcForest has been widely used in various applications. However, the current multi-grained scanning of gcForest produces many redundant feature vectors, and this increases the time cost of the model. To screen out redundant feature vectors, we introduce a hashing screening mechanism for multi-grained scanning and propose a model called HW-Forest which adopts two strategies, hashing screening and window screening. HW-Forest employs perceptual hashing algorithm to calculate the similarity between feature vectors in hashing screening strategy, which is used to remove the redundant feature vectors produced by multi-grained scanning and can significantly decrease the time cost and memory consumption. Furthermore, we adopt a self-adaptive instance screening strategy to improve the performance of our approach, called window screening, which can achieve higher accuracy without hyperparameter tuning on different datasets. Our experimental results show that HW-Forest has higher accuracy than other models, and the time cost is also reduced. Copyright © 2022, The Authors. All rights reserved.

关键词： Scanning

VSRQ: Quantitative Assessment Method for Safety Risk of Vehicle Intelligent Connected System

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Tian Guan, Wenshan Miao, Hao Huang, Xiujie Liu, Zhiquan Wang, Chaonan Guan, Quanlong Fang, Liangda Duan, Zhifei Department of Cyberspace Security College of Information Science and Technology Jinan University Guangzhou511486 China Department of Computer Science College of Information Science and Technology Jinan University Guangzhou510632 China College of Information Science and Technology Jinan University Guangdong Institution of Smart Education Jinan University Guangzhou510632 China College of Cyber Security Jinan University Guangzhou510632 China Guangdong Gene Data Processing and Analysis Engineering Research Center China Pazhou Laboratory Guangzhou510330 China Guangxi Key Laboratory of Trusted Software Guilin University of Electronic Technology Guilin541004 China Guangzhou XPeng Motors Technology Co. Ltd No.8 Songgang Road Changxing Street Cencun Tianhe District Guangzhou China

The field of intelligent connected in modern vehicles continues to expand, and the functions of vehicles become more and more complex with the development of the times. This has also led to an increasing number of vehicle vulnerabilities and many safety issues. Therefore, it is particularly important to identify high-risk vehicle intelligent connected systems, because it can inform security personnel which systems are most vulnerable to attacks, allowing them to conduct more thorough inspections and tests. In this paper, we develop a new model for vehicle risk assessment by combining I-FAHP with FCA clustering: VSRQ model. We extract important indicators related to vehicle safety, use fuzzy cluster analys (FCA) combined with fuzzy analytic hierarchy process (FAHP) to mine the vulnerable components of the vehicle intelligent connected system, and conduct priority testing on vulnerable components to reduce risks and ensure vehicle safety. We evaluate the model on OpenPilot and experimentally demonstrate the effectiveness of the VSRQ model in identifying the safety of vehicle intelligent connected systems. The experiment fully complies with ISO 26262 and ISO/SAE 21434 standards, and our model has a higher accuracy rate than other models. These results provide a promising new research direction for predicting the security risks of vehicle intelligent connected systems and provide typical application tasks for VSRQ. The experimental results show that the accuracy rate is 94.36%, and the recall rate is 73.43%, which is at least 14.63% higher than all other known indicators. Copyright © 2023, The Authors. All rights reserved.

关键词： Vehicle safety

Revisiting smoothed online learning 21

学校读者我要写书评

暂无评论

Revisiting smoothed online learning

Proceedings of the 35th International Conference on Neural Information Processing Systems

作者： Lijun Zhang Wei Jiang Shiyin Lu Tianbao Yang National Key Laboratory for Novel Software Technology Nanjing University Nanjing China and Peng Cheng Laboratory Shenzhen Guangdong China National Key Laboratory for Novel Software Technology Nanjing University Nanjing China Department of Computer Science The University of Iowa Iowa City IA

ISBN: (纸本)9781713845393

In this paper, we revisit the problem of smoothed online learning, in which the online learner suffers both a hitting cost and a switching cost, and target two performance metrics: competitive ratio and dynamic regret with switching cost. To bound the competitive ratio, we assume the hitting cost is known to the learner in each round, and investigate the simple idea of balancing the two costs by an optimization problem. Surprisingly, we find that minimizing the hitting cost alone is max(1, 2/α) -competitive for α-polyhedral functions and 1 + 4/λ-competitive for λ-quadratic growth functions, both of which improve state-of-the-art results significantly. Moreover, when the hitting cost is both convex and λ-quadratic growth, we reduce the competitive ratio to 1 + 2/√λ by minimizing the weighted sum of the hitting cost and the switching cost. To bound the dynamic regret with switching cost, we follow the standard setting of online convex optimization, in which the hitting cost is convex but hidden from the learner before making predictions. We modify Ader, an existing algorithm designed for dynamic regret, slightly to take into account the switching cost when measuring the performance. The proposed algorithm, named as Smoothed Ader, attains an optimal $O(\sqrt{T(1+P_T)})$ bound for dynamic regret with switching cost, where PT is the path-length of the comparator sequence. Furthermore, if the hitting cost is accessible in the beginning of each round, we obtain a similar guarantee without the bounded gradient condition, and establish an $\Omega(\sqrt{T(1+P_T)})$ lower bound to confirm the optimality.

关键词：

CLDG: Contrastive Learning on Dynamic Graphs

学校读者我要写书评

暂无评论

CLDG: Contrastive Learning on Dynamic Graphs

International Conference on Data Engineering

作者： Yiming Xu Bin Shi Teng Ma Bo Dong Haoyi Zhou Qinghua Zheng Department of Computer Science and Technology Xi’an Jiaotong University China Shaanxi Provincial Key Laboratory of Big Data Knowledge Engineering Xi’an Jiaotong University China Department of Distance Education Xi’an Jiaotong University China School of Software Beihang University China Advanced Innovation Center for Big Data and Brain Computing Beihang University China

The graph with complex annotations is the most potent data type, whose constantly evolving motivates further exploration of the unsupervised dynamic graph representation. One of the representative paradigms is graph contrastive learning. It constructs self-supervised signals by maximizing the mutual information between the statistic graph’s augmentation views. However, the semantics and labels may change within the augmentation process, causing a significant performance drop in downstream tasks. This drawback becomes greatly magnified on dynamic graphs. To address this problem, we designed a simple yet effective framework named CLDG. Firstly, we elaborate that dynamic graphs have temporal translation invariance at different levels. Then, we proposed a sampling layer to extract the temporally-persistent signals. It will encourage the node to maintain consistent local and global representations, i.e., temporal translation invariance under the timespan views. The extensive experiments demonstrate the effectiveness and efficiency of the method on seven datasets by outperforming eight unsupervised state-of-the-art baselines and showing competitiveness against four semi-supervised methods. Compared with the existing dynamic graph method, the number of model parameters and training time is reduced by an average of 2,001.86 times and 130.31 times on seven datasets, respectively. The code and data are available at: https://***/yimingxu24/CLDG.

关键词：

ACL-Net: Adaptive and Collaborative Learning Network for Multi-Site Prostate MRI Segmentation

学校读者我要写书评

暂无评论

ACL-Net: Adaptive and Collaborative Learning Network for Mul...

IEEE International Conference on Big Data

作者： Zibo Ma Bo Zhang Zheng Zhang Wendong Wang Yue Mi Haiwen Huang Jingyun Wu Beijing University of Posts and Telecommunications Beijing China State Key Laboratory of Networking and Switching Technology School of Computer Science (National Pilot Software Engineering School) Beijing University of Posts and Telecommunications School of Modern Post Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology School of Computer Science (National Pilot Software Engineering School) Beijing University of Posts and Telecommunications Beijing China Peking University First Hospital Beijing China Department of Urology Peking University First Hospital Department of Urology Peking University First Hospital Beijing China Department of Radiology Peking University First Hospital

ISBN: (纸本)9781665480468

High-performance deep learning models require large amounts of data with high quality annotations for model training, while the labeling work usually takes a lot of time for the experts. Meanwhile, the inter-observer variability al-ways exist between annotations from different experts and the distribution shift between the data acquired from different medical institutions. To address these challenges, we propose an end-to-end domain adaptive collaborative learning network for multi-institutional prostate MRI segmentation. Specifically, we introduce an unpaired image translation module to match the image domains between different institutions, which can alleviate the heterogeneity between 1.5T and 3T prostate MR images during model training. Moreover, we design a self-taught strategy to transfer domain-aware knowledge to jointly learn generic and unique representations. Furthermore, we evaluate our approach in scenarios with limited or without annotations, experimental results show that our approach has better adaptation performance than traditional supervised learning approaches, and has the potential to extend to unsupervised domain adaptation scenario. We also evaluate our approach with prostate MRI segmentation benchmark datasets, experimental results show that our approach outperforms several state-of-the-art methods.

关键词： Training Deep learning Image segmentation Adaptation models Adaptive systems Annotations Federated learning

UAV Swarm-enabled Collaborative Secure Relay Communications with Time-domain Colluding Eavesdropper

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Chuang Sun, Geng Wu, Qingqing Li, Jiahui Liang, Shuang Niyato, Dusit Leung, Victor C.M. The College of Computer Science and Technology Jilin University Changchun130012 China The Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Changchun130012 China The Department of Electronic Engineering Shanghai Jiao Tong University Shanghai China Pillar of Engineering Systems and Design Singapore University of Technology and Design 487372 Singapore The School of Information Science and Technology Northeast Normal University Changchun130024 China The School of Computer Science and Engineering Nanyang Technological University 639798 Singapore The College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China The Department of Electrical and Computer Engineering University of British Columbia VancouverBCV6T 1Z4 Canada

Unmanned aerial vehicles (UAVs) as aerial relays are practically appealing for assisting Internet of Things (IoT) network. In this work, we aim to utilize the UAV swarm to assist the secure communication between the micro base station (MBS) equipped with the planar array antenna (PAA) and the IoT terminal devices by collaborative beamforming (CB), so as to counteract the effects of collusive eavesdropping attacks in time-domain. Specifically, we formulate a UAV swarm-enabled secure relay multi-objective optimization problem (US2RMOP) for simultaneously maximizing the achievable sum rate of associated IoT terminal devices, minimizing the achievable sum rate of the eavesdropper and minimizing the energy consumption of UAV swarm, by jointly optimizing the excitation current weights of both MBS and UAV swarm, the selection of the UAV receiver, the position of UAVs and user association order of IoT terminal devices. Furthermore, the formulated US2RMOP is proved to be a non-convex, NP-hard and large-scale optimization problem. Therefore, we propose an improved multi-objective grasshopper algorithm (IMOGOA) with some specific designs to address the problem. Simulation results exhibit the effectiveness of the proposed UAV swarm-enabled collaborative secure relay strategy and demonstrate the superiority of IMOGOA. Copyright © 2023, The Authors. All rights reserved.

关键词： Multiobjective optimization

QoS-aware flow scheduling for energy-efficient cloud data centre network

学校读者我要写书评

暂无评论

QoS-aware flow scheduling for energy-efficient cloud data ce...

作者： Wang, Songyun Yuan, Jiabin Zhang, Xiaoda Qian, Zhuzhong Li, Xin You, Ilsun College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing211106 China State Key Laboratory for Novel Software Technology Nanjing University Nanjing210023 China CCST Nanjing University of Aeronautics and Astronautics State Key Laboratory for Novel Software Technology Nanjing University Nanjing210000 China Department of Information Security Engineering Soonchunhyang University Asan-si Chungcheongnam-do Korea Republic of

It is highly valuable to achieve energy-efficient cloud data centres, which always act as the basic infrastructures. This paper thus aims at reducing the energy consumption of network devices in cloud data centres by flexible flow scheduling. For such an aim, it is necessary to guarantee the flow-level performance, which is a critical requirement for quality of service (QoS) in production data centres. Hence, this paper takes both energy reduction and QoS into account for flow scheduling, and then proposes a three-phase framework to control the energy consumption for cloud data centre network (DCN) while guaranteeing flow-level performance. The core idea is to turn off the unnecessary switches off. From extensive simulations, it is shown that our solution could not only reduce about 20% of energy on average than the case with all switches on, but also maintain good flow-level performance, stability and fault tolerance simultaneously. Copyright © 2020 Inderscience Enterprises Ltd.

关键词： Quality of service