检索结果-内蒙古大学图书馆

Mining Latent Classes for Few-shot Segmentation

学校读者我要写书评

暂无评论

Mining Latent Classes for Few-shot Segmentation

International Conference on computer Vision (ICCV)

作者： Lihe Yang Wei Zhuo Lei Qi Yinghuan Shi Yang Gao State Key Laboratory for Novel Software Technology Nanjing University Tencent Key Lab of Computer Network and Information Integration (Ministry of Education) Southeast University

ISBN: (纸本)9781665428132

Few-shot segmentation (FSS) aims to segment unseen classes given only a few annotated samples. Existing methods suffer the problem of feature undermining, i.e., potential novel classes are treated as background during training phase. Our method aims to alleviate this problem and enhance the feature embedding on latent novel classes. In our work, we propose a novel joint-training framework. Based on conventional episodic training on support-query pairs, we introduce an additional mining branch that exploits latent novel classes via transferable sub-clusters, and a new rectification technique on both background and fore-ground categories to enforce more stable prototypes. Over and above that, our transferable sub-cluster has the ability to leverage extra unlab.led data for further feature enhancement. Extensive experiments on two FSS benchmarks demonstrate that our method outperforms previous state-of-the-art by a large margin of 3.7% mIOU on PASCAL-5 i and 7.0% mIOU on COCO-20 i at the cost of 74% fewer parameters and 2.5x faster inference speed. The source code is availab.e at https://***/LiheYoung/MiningFSS.

关键词： Training computer vision Costs Codes Training data Prototypes Benchmark testing

Disentangled High Quality Salient Object Detection

学校读者我要写书评

暂无评论

Disentangled High Quality Salient Object Detection

International Conference on computer Vision (ICCV)

作者： Lv Tang Bo Li Yijie Zhong Shouhong Ding Mofei Song Youtu Lab Tencent Shanghai China The School of Computer Science and Engineering Southeast University Nanjing China The Key Lab of Computer Network and Information Integration (Ministry of Education) Southeast University Nanjing China

ISBN: (纸本)9781665428132

Aiming at discovering and locating most distinctive objects from visual scenes, salient object detection (SOD) plays an essential role in various computer vision systems. Coming to the era of high resolution, SOD methods are facing new challenges. The major limitation of previous methods is that they try to identify the salient regions and estimate the accurate objects boundaries simultaneously with a single regression task at low-resolution. This practice ignores the inherent difference between the two difficult problems, resulting in poor detection quality. In this paper, we propose a novel deep learning framework for high-resolution SOD task, which disentangles the task into a low-resolution saliency classification network (LRSCN) and a high-resolution refinement network (HRRN). As a pixel-wise classification task, LRSCN is designed to capture sufficient semantics at low-resolution to identify the definite salient, background and uncertain image regions. HRRN is a regression task, which aims at accurately refining the saliency value of pixels in the uncertain region to preserve a clear object boundary at high-resolution with limited GPU memory. It is worth noting that by introducing uncertainty into the training process, our HRRN can well address the high-resolution refinement task without using any high-resolution training data. Extensive experiments on high-resolution saliency datasets as well as some widely used saliency benchmarks show that the proposed method achieves superior performance compared to the state-of-the-art methods.

关键词： Training Deep learning computer vision Uncertainty Semantics Refining Graphics processing units

A Lightening Fine-grained Access Control Scheme Supporting Full Policy Hiding in Cloud Storage

学校读者我要写书评

暂无评论

A Lightening Fine-grained Access Control Scheme Supporting F...

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Jiaoli Shi Chao Hu Shimao Yao Zhuolin Mei Bin Wu Hui Li School of Computer and Big Data Jiujiang Univ. Jiujiang China School of Software IoT Engineering Jiangxi Univ. of Finance and Economics Nanchang China Jiujiang Key Lab. of Network and Information Security Jiujiang China School of Computer and Big Data Jiujiang University Jiujiang China School of Elec. and Comp. Engineering Peking University Shenzhen China

Efficiency, security, and flexibility are difficult to balance in a full policy hiding CP-ABE scheme. This paper creatively converts the privacy matching problem of access policy and a user’s private key to the privacy three-party set computation problem. Then the new privacy set computation problem is resolved by a message driven key generation method. Additionally, the ROBDD (Reduced Ordered Binary Decision Diagrams) is utilized to rich access structure expression and convert access policies into feasible path sets. Each feasible path is composed of attributes and these attributes are bound to the ciphertext using CP-ABE. Based on the innovative ideas mentioned above, we present a Lighten Fine-grained Access Control scheme supporting Full Policy Hidden (LFAC-FPH) that integrates data confidentiality protection, access policy rich expression but complete hiding, fine-grained access control, and lightweight terminal computing. Our scheme has three advantages: 1) High Efficiency. The proposed scheme can achieve fast privacy matching of access policies during user decryption. 2) Full Policy Hiding. The access policy bound by an owner on the ciphertext does not leak any information. Moreover, a user cannot know attributes they do not own. 3) High Flexibility. The decryption process does not require online from owners or KGC, and system attributes can be added at any time.

关键词：

MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency Regularization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Duan, Yue Zhao, Zhen Qi, Lei Wang, Lei Zhou, Luping Shi, Yinghuan Gao, Yang The National Key Laboratory for Novel Software Technology The National Institute of Healthcare Data Science Nanjing University Nanjing210023 China The School of Electrical and Information Engineering The University of Sydney SydneyNSW2006 Australia The School of Computer Science and Engineering The Key Lab of Computer Network and Information Integration Ministry of Education Southeast University Nanjing211189 China The School of Computing and Information Technology University of Wollongong WollongongNSW2522 Australia

The core issue in semi-supervised learning (SSL) lies in how to effectively leverage unlab.led data, whereas most existing methods tend to put a great emphasis on the utilization of high-confidence samples yet seldom fully explore the usage of low-confidence samples. In this paper, we aim to utilize low-confidence samples in a novel way with our proposed mutex-based consistency regularization, namely MutexMatch. Specifically, the high-confidence samples are required to exactly predict "what it is" by conventional True-Positive Classifier, while low-confidence samples are employed to achieve a simpler goal — to predict with ease "what it is not" by True-Negative Classifier. In this sense, we not only mitigate the pseudo-lab.ling errors but also make full use of the low-confidence unlab.led data by consistency of dissimilarity degree. MutexMatch achieves superior performance on multiple benchmark datasets, i.e., CIFAR-10, CIFAR-100, SVHN, STL-10, mini-ImageNet and Tiny-ImageNet. More importantly, our method further shows superiority when the amount of lab.led data is scarce, e.g., 92.23% accuracy with only 20 lab.led data on CIFAR-10. Our code and model weights have been released at https://***/NJUyued/MutexMatch4SSL. Copyright © 2022, The Authors. All rights reserved.

关键词： Benchmarking

DRL-Based Optimization for AoI and Energy Consumption in C-V2X Enabled IoV

学校读者我要写书评

暂无评论

IEEE Transactions on Green Communications and Networking

IEEE Transactions on Green Communications and networking 2025年

作者： Zhang, Zheng Wu, Qiong Fan, Pingyi Cheng, Nan Chen, Wen Letaief, Khaled B. Jiangnan University School of Internet of Things Engineering Wuxi214122 China Tsinghua University Beijing National Research Center for Information Science and Technology Department of Electronic Engineering State Key laboratory of Space Network and Communications Tsinghua University Beijing100084 China Xidian University State Key Lab. of ISN School of Telecom-munications Engineering Xi'an710071 China Shanghai Jiao Tong University Department of Electronic Engineering Shanghai200240 China Department of Electrical and Computer Engineer-ing Hong Kong

To address communication latency issues, the Third Generation Partnership Project (3GPP) has defined Cellular-Vehicle to Everything (C-V2X) technology, which includes Vehicle-to-Vehicle (V2V) communication for direct vehicle-to-vehicle communication. However, this method requires vehicles to autonomously select communication resources based on the Semi-Persistent Scheduling (SPS) protocol, which may lead to collisions due to different vehicles sharing the same communication resources, thereby affecting communication effectiveness. Non-Orthogonal Multiple Access (NOMA) is considered a potential solution for handling large-scale vehicle communication, as it can enhance the Signal-to-Interference-plus-Noise Ratio (SINR) by employing Successive Interference Cancellation (SIC), thereby reducing the negative impact of communication collisions. When evaluating vehicle communication performance, traditional metrics such as reliability and transmission delay present certain contradictions. Introducing the new metric Age of information (AoI) provides a more comprehensive evaluation of communication system. Additionally, to ensure service quality, user terminals need to possess high computational capabilities, which may lead to increased energy consumption, necessitating a trade-off between communication energy consumption and effectiveness. Given the complexity and dynamics of communication systems, Deep Reinforcement Learning (DRL) serves as an intelligent learning method capable of learning optimal strategies in dynamic environments. Therefore, this paper analyzes the effects of multi-priority queues and NOMA on AoI in the C-V2X vehicular communication system and proposes an energy consumption and AoI optimization method based on DRL. Finally, through comparative simulations with baseline methods, the proposed approach demonstrates its advances in terms of energy consumption and AoI. © 2017 IEEE.

关键词： Deep reinforcement learning

Veca: A High-Performance Consensus Algorithm for State Machine Replication 4th

学校读者我要写书评

暂无评论

Veca: A High-Performance Consensus Algorithm for State Machi...

4th International Conference on Human Centered Computing, HCC 2018

作者： Yan, Xiangguang Wang, Yun Southeast University Nanjing China Key Lab of Computer Network and Information Integration MOE Nanjing China

ISBN: (纸本)9783030151263

Consensus algorithm is fundamental for distributed systems. It requires high throughput, low latency and high availab.lity. However, to the best of our knowledge, these requirements cannot be satisfied well at the same time in any existing consensus algorithm. In this paper, we propose Veca, a consensus algorithm for state machine replication that tries to satisfy the requirements at the same time as well as possible. Veca is a leaderless consensus algorithm for which all replicas can commit commands concurrently at any time, and each command can be committed after just one round of communication with a majority of replicas in the normal case. Veca separates agreement from ordering and execution, which allows all replicas to commit commands concurrently without determining their order, but to track their dependencies using vector clocks. Then a subsequent replay phase assigns an order to the commands and executes them in that order. Commands are committed out of order and then be executed in the same order by all replicas. A replica can take the initiative to learn the decision for an instance using a failure recovery protocol. The leaderless design makes the systems built with Veca provide continuous service as long as more than half of the replicas are availab.e. We show that Veca has higher throughput, lower latency and higher availab.lity than several typical consensus algorithms. The correctness of Veca has been proved theoretically and its advantages are demonstrated through an experimental evaluation in LAN and WAN. © 2019, Springer Nature Switzerland AG.

关键词： computers

Hybrid Attentional Memory network for Computational drug repositioning

学校读者我要写书评

暂无评论

arXiv 2020年

作者： He, Jieyue Yang, Xinxing Gong, Zhuo Zamit, Lbrahim School of Computer Science and Engineering Key Lab of Computer Network & Information Integration MOE Southeast University Nanjing210018 China

Background: Drug repositioning is designed to discover new uses of known drugs, which is an important and efficient method of drug discovery. Researchers only use one certain type of Collab.rative Filtering (CF) models for drug repositioning currently, like the neighborhood based approaches which are good at mining the local information contained in few strong drug-disease associations, or the latent factor based models which are effectively capture the global information shared by a majority of drug-disease associations. Few researchers have combined these two types of CF models to derive a hybrid model with the advantages of both of them. Besides, the cold start problem has always been a major challenge in the field of computational drug repositioning, which restricts the inference ability of relevant models. Results: Inspired by the memory network, we propose the Hybrid Attentional Memory network (HAMN) model, a deep architecture combines two classes of CF model in a nonlinear manner. Firstly, the memory unit and the attention mechanism are combined to generate the neighborhood contribution representation to capture the local structure of few strong drug-disease associations. Then a variant version of the autoencoder is used to extract the latent factor of drugs and diseases to capture the overall information shared by a majority of drug-disease associations. In that process, ancillary information of drugs and diseases can help to alleviate the cold start problem. Finally, in the prediction stage, the neighborhood contribution representation is combined with the drug latent factor and disease latent factor to produce the predicted value. Comprehensive experimental results on two real data sets show that our proposed HAMN model is superior to other comparison models according to the AUC, AUPR and HR indicators. Conclusions: Through the performance on two real-world data sets, we believe that the proposed model can effectively alleviate the cold start problem in th

关键词： Collab.rative filtering

Better Pseudo-lab.l: Joint Domain-aware lab.l and Dual-classifier for Semi-supervised Domain Generalization

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Wang, Ruiqi Qi, Lei Shi, Yinghuan Gao, Yang State Key Laboratory for Novel Software Technology Nanjing University Nanjing China National Institute of Healthcare Data Science Nanjing University Nanjing China School of Computer Science and Engineering Key Lab of Computer Network and Information Integration Ministry of Education Southeast University Nanjing China

With the goal of directly generalizing trained model to unseen target domains, domain generalization (DG), a newly proposed learning paradigm, has attracted considerable attention. Previous DG models usually require a sufficient quantity of annotated samples from observed source domains during training. In this paper, we relax this requirement about full annotation and investigate semisupervised domain generalization (SSDG) where only one source domain is fully annotated along with the other domains totally unlab.led in the training process. With the challenges of tackling the domain gap between observed source domains and predicting unseen target domains, we propose a novel deep framework via joint domain-aware lab.ls and dual-classifier to produce high-quality pseudo-lab.ls. Concretely, to predict accurate pseudo-lab.ls under domain shift, a domain-aware pseudo-lab.ling module is developed. Also, considering inconsistent goals between generalization and pseudo-lab.ling: former prevents overfitting on all source domains while latter might overfit the unlab.led source domains for high accuracy, we employ a dual-classifier to independently perform pseudo-lab.ling and domain generalization in the training process. When accurate pseudo-lab.ls are generated for unlab.led source domains, the domain mixup operation is applied to augment new domains between lab.led and unlab.led domains, which is beneficial for boosting the generalization capability of the model. Extensive results on publicly availab.e DG benchmark datasets show the efficacy of our proposed SSDG method. Copyright © 2021, The Authors. All rights reserved.

关键词： Image recognition

Robust Generalization against Photon-Limited Corruptions via Worst-Case Sharpness Minimization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Huang, Zhuo Zhu, Miaoxi Xia, Xiaobo Shen, Li Yu, Jun Gong, Chen Han, Bo Du, Bo Liu, Tongliang Syndey Ai Centre The University of Sydney Australia National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University China Jd Explore Academy China Department of Automation University of Science and Technology of China China Key Lab. of Intelligent Percept. and Syst. for High-Dimensional Information of Ministry of Education School of Computer Science and Engineering Nanjing University of Science and Technology China Department of Computer Science Hong Kong Baptist University Hong Kong

Robust generalization aims to tackle the most challenging data distributions which are rare in the training set and contain severe noises, i.e., photon-limited corruptions. Common solutions such as distributionally robust optimization (DRO) focus on the worst-case empirical risk to ensure low training error on the uncommon noisy distributions. However, due to the over-parameterized model being optimized on scarce worst-case data, DRO fails to produce a smooth loss landscape, thus struggling on generalizing well to the test set. Therefore, instead of focusing on the worst-case risk minimization, we propose SharpDRO by penalizing the sharpness of the worst-case distribution, which measures the loss changes around the neighbor of learning parameters. Through worst-case sharpness minimization, the proposed method successfully produces a flat loss curve on the corrupted distributions, thus achieving robust generalization. Moreover, by considering whether the distribution annotation is availab.e, we apply SharpDRO to two problem settings and design a worst-case selection process for robust generalization. Theoretically, we show that SharpDRO has a great convergence guarantee. Experimentally, we simulate photon-limited corruptions using CIFAR10/100 and ImageNet30 datasets and show that SharpDRO exhibits a strong generalization ability against severe corruptions and exceeds well-known baseline methods with large performance gains. Copyright © 2023, The Authors. All rights reserved.

关键词： Optimization