检索结果-内蒙古大学图书馆

Artificial Intelligence technology (ACAIT), Asian Conference on

作者： Jialei Wang Qihao Xu Xiaoling Luo Yong Xu Department of Computer Science and Technology Harbin Institute of Technology Shenzhen China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Guangdong Provincial Key Laboratory of Novel Security Intelligence Technologies Shenzhen China

ISBN: (数字)9798331517090

ISBN: (纸本)9798331517106

With the aging of population and the change of people’s lifestyle, fundus diseases have become the main irreversible cause of blindness. computer-aided classification of multiple fundus diseases can improve the efficiency of diagnosis and prevent the further deterioration of fundus diseases. However, the classification approach faces great challenges due to the limited available data, complex types of fundus diseases, and subtle lesion features. To solve the above problems, we propose a neural network using multi-channel mutual learning mechanism. In this work, we propose Attention-Augmentation Block, which adjusts the learning focus of different channels by disease-grouping training on samples to make full use of the limited data. Furthermore, to enhance the feature representation capability of the model, the proposed method implements the mutual enhancement and fusion of multi-channel features through the mutual learning mechanism by designing the Multi-Channel Mutual Learning Unit. Experimental results show that our proposed method can achieve better classification performance of multiple fundus diseases than the state-of-the-art methods. The code of our paper is available in the attachment.

关键词： Learning systems Training Neural networks Multi label classification Feature extraction Medical diagnosis Lesions Medical diagnostic imaging Diseases Image classification

来源：评论

学校读者我要写书评

暂无评论

Planning Paths for UAVs to Collect Data from Disconnected Sensor Networks 27

Planning Paths for UAVs to Collect Data from Disconnected Se...

引用

27th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2021

作者： Li, Xuemeng Zhu, Xiaojun Dong, Chao College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics China State Key Laboratory of Novel Software Technology Nanjing University China

ISBN: (纸本)9781665408783

In case of energy depletion at some nodes, a sensor network may become disconnected, and data from live nodes may not be sent to the sink. This problem can be solved by a UAV acting as a mobile data collector. However, the traditional approach is to treat live sensor nodes as independent data sources. In this paper, we consider a cooperative approach where live sensors form sub-networks, instead of acting indepen-dently. Sensors within a subnetwork can send data to each other. A UAV only needs to visit subnetworks, instead of individual sensor nodes, to collect data. We formulate the problem and propose an iterative optimization algorithm to find the shortest data collection trajectory. Given any trajectory, our algorithm iteratively merges nodes within the same subnetwork and adjusts hovering locations to minimize trajectory length. Simulations show that, compared with the traditional TSP formulation, our approach can significantly reduces the length of the trajectory. © 2021 IEEE.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Nerftap: Enhancing Transferability of Adversarial Patches on Face Recognition Using Neural Radiance Fields

SSRN

引用

SSRN 2024年

作者： Liu, Xiaoliang Shen, Fu-Rao Han, Feng Zhao, Jian Nie, Changhai State Key Laboratory for Novel Software Technology Nanjing University China Department of Computer Science and Technology Nanjing University China School of Artificial Intelligence Nanjing University China School of Electronic Science and Engineering Nanjing University China

Face recognition (FR) technology plays a crucial role in various applications, but its vulnerability to adversarial attacks poses significant security concerns. Existing research primarily focuses on transferability to different FR models, overlooking the direct transferability to victim's face images, which is a practical threat in real-world scenarios. In this study, we propose a novel adversarial attack method that considers both the transferability to the FR model and the victim's face image, called NerfTAP. Leveraging Neural Radiance Fields (NeRF)-based 3D Generative Adversarial Network (GAN), we generate new view face images for the source and target subjects to enhance transferability of adversarial patches. We introduce a style consistency loss to ensure the visual similarity between the adversarial UV map and the target UV map under a 0-1 mask, enhancing the effectiveness and naturalness of the generated adversarial face images. Extensive experiments and evaluations on various FR models demonstrate the superiority of our approach over existing attack techniques. Our work provides valuable insights for enhancing the robustness of FR systems in practical adversarial settings. © 2024, The Authors. All rights reserved.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Energy-guided diffusion sampling for offline-to-online reinforcement learning 24

Energy-guided diffusion sampling for offline-to-online reinf...

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Xu-Hui Liu Tian-Shuo Liu Shengyi Jiang Ruifeng Chen Zhilong Zhang Xinwei Chen Yang Yu National Key Laboratory for Novel Software Technology & School of Artificial Intelligence Nanjing University China National Key Laboratory for Novel Software Technology & School of Artificial Intelligence Nanjing University China and Polixir Technologies Department of Computer Science The University of Hong Kong Hong Kong China Polixir Technologies

Combining offline and online reinforcement learning (RL) techniques is indeed crucial for achieving efficient and safe learning where data acquisition is expensive. Existing methods replay offline data directly in the online phase, resulting in a significant challenge of data distribution shift and subsequently causing inefficiency in online fine-tuning. To address this issue, we introduce an innovative approach, Energy-guided DIffusion Sampling (EDIS), which utilizes a diffusion model to extract prior knowledge from the offline dataset and employs energy functions to distill this knowledge for enhanced data generation in the online phase. The theoretical analysis demonstrates that EDIS exhibits reduced suboptimality compared to solely utilizing online data or directly reusing offline data. EDIS is a plug-in approach and can be combined with existing methods in offline-to-online RL setting. By implementing EDIS to off-the-shelf methods Cal-QL and IQL, we observe a notable 20% average improvement in empirical performance on MuJoCo, AntMaze, and Adroit environments. Code is available at https://***/liuxhym/EDIS

关键词：

来源：评论

学校读者我要写书评

暂无评论

Exact Algorithm for Batch Scheduling with Many-to-Many Job-Tool Matching and Job Release Time Constraints

引用

IEEE Transactions on Automation science and Engineering 2025年 22卷 16499-16511页

作者： Li, Shengchao Liu, Shixin Zhao, Ziyan Zhou, Mengchu Northeastern University State Key Laboratory of Synthetical Automation for Process Industries College of Information Science and Engineering Shenyang 110819 China Northeastern University at Qinhuangdao Hebei Key Laboratory of Industrial Software for Advanced Planning and Scheduling of Complex Industrial Processes School of Control Engineering Qinhuangdao 066004 China New Jersey Institute of Technology Department of Electrical and Computer Engineering Newark 07102 NJ United States

Job-dependent tool switching is necessary in many batch processing systems (BPSs). Heterogeneous tool demand and extra time consumption for tool switch bring great challenge for high-performance production scheduling in BPSs. In this study, we present a novel batch scheduling problem derived from a satellite vibration test system. Different from basic scheduling problems in BPSs with tool switching, it has complications caused by many-to-many task-tool matching and job release time constraints. Specifically, each job has predetermined candidate tools. One of them should be selected and equipped to process a given job. Unlike sequence-dependent setup time that is only related to two adjacent jobs, tool switching time is additionally dependent on the tools selected by both jobs from their respective candidate tools. Both job sequence and tool selections need to be determined to minimize makespan. Mixed-integer linear programming (MILP) and constraint programming (CP) models are developed for formulating the problem. We prove that the problem is NP hard. To obtain exact solutions, we propose a novel branch-and-bound algorithm (B&B) based on the analysis and extraction of problem properties. It integrates a domain reduction mechanism, branching strategies, and dominance rules. Among them, the domain reduction mechanism reduces the number of jobs to be scheduled;the branching strategies accelerate solving speed;and the dominance rules save the memory consumption. Ablation studies and key influencing factor experiments are performed. The results show that the proposed B&B can optimally solve much larger instances than state-of-the-art methods (including some heuristics). It can optimally solve industrial-size problems in a short time, implying that it can well satisfy the performance demand of a satellite vibration test system and thus advancing the filed of optimal scheduling for BPSs. © 2004-2012 IEEE.

关键词： Batch scheduling branch-and-bound algorithm heterogenerous tool demand mixed-integer linear programming satellite vibration test system

来源：评论

学校读者我要写书评

暂无评论

Decentralized Counterfactual Value with Threat Detection in Multi-Agent Mixed Cooperative and Competitive Environment

SSRN

引用

SSRN 2023年

作者： Dong, Shaokang Li, Chao Yang, Shangdong Gao, Yang State Key Laboratory for Novel Software Technology Nanjing University China School of Computer Science Nanjing University of Posts and Telecommunications China

This paper introduces a fully decentralized method to address the problem of general mixed cooperation and competition in multi-agent setting. There are two main types of traditional Multi-Agent Reinforcement Learning (MARL) methods. The first kind is communication-based method which is designed to achieve coordination to deal with the non-stationary problem. However, it causes great communication overhead and can only be applied in fully cooperative environment. Another kind of methods follows Centralized Training and Decentralized Execution (CTDE) framework, which needs a centralized critic in the training process resulting in the huge computational complexity as the number of agents grows. In addition, neither approach is fully decentralized, because they either need to communicate the implicit information, or need to know the rewards of other agents. Different to the above methods, this paper introduces a Decentralized Counterfactual Value (DCV) to model other agents' behaviors, combined with the Threat Detection mechanism to learn the latent competitive or cooperative relationship with others. Finally, we prove that our method DCVTD does not affect the convergence of the basic algorithm and validate that DCVTD outperforms other fully decentralized methods, CTDE methods and other agent modeling or social influence methods in the experiments. © 2023, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Momentum Posterior Regularization for Multi-hop Dense Retrieval

arXiv

引用

arXiv 2024年

作者： Xia, Zehua Wu, Yuyang Xia, Yiyun Nguyen, Cam-Tu State Key Laboratory for Novel Software Technology Nanjing University China School of Computer Science Nanjing University China School of Artificial Intelligence Nanjing University China

Multi-hop question answering (QA) often requires sequential retrieval (multi-hop retrieval), where each hop retrieves missing knowledge based on information from previous hops. To facilitate more effective retrieval, we aim to distill knowledge from a posterior retrieval, which has access to posterior information like an answer, into a prior retrieval used during inference when such information is unavailable. Unfortunately, current methods for knowledge distillation in one-time retrieval are ineffective for multi-hop QA due to two issues: 1) Posterior information is often defined as the response (i.e. the answer), which may not clearly connect to the query without intermediate retrieval;and 2) The large knowledge gap between prior and posterior retrievals makes existing distillation methods unstable, even resulting in performance loss. As such, we propose MoPo (Momentum Posterior Regularization) with two key innovations: 1) Posterior information of one hop is defined as a query-focus summary from the golden knowledge of the previous and current hops;2) We develop an effective training strategy where the posterior retrieval is updated along with the prior retrieval via momentum moving average method, allowing smoother and effective distillation. Experiments on HotpotQA and StrategyQA demonstrate that MoPo outperforms existing baselines in both retrieval and downstream QA tasks‡ © 2024, CC BY-NC-SA.

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

ATOM: Automated Black-Box Testing of Multi-Label Image Classification Systems

ATOM: Automated Black-Box Testing of Multi-Label Image Class...

引用

IEEE International Conference on Automated software Engineering (ASE)

作者： Shengyou Hu Huayao Wu Peng Wang Jing Chang Yongjun Tu Xiu Jiang Xintao Niu Changhai Nie Department of Computer Science and Technology State Key Laboratory for Novel Software Technology Nanjing University China Guangdong OPPO Mobile Telecommunications Corp. Ltd. China

Multi-label Image Classification Systems (MICSs) developed based on Deep Neural Networks (DNNs) are extensively used in people's daily life. Currently, although there are a variety of approaches to test DNN-based systems, they typically rely on the internals of DNNs to design test cases, and do not take the core specification of MICS (i.e., correctly recognizing multiple objects in a given image) into account. In this paper, we propose ATOM, an automated and systematic black-box testing framework for testing MICS. Specifically, ATOM exploits the label combination as the testing adequacy criteria, hoping to systematically examine the impact of correlations between a fixed number of labels on the classification ability of MICS. Then, ATOM leverages image search engine and natural language processing to find test images that are not only common to the real-world, but also relevant to target label combinations. Finally, ATOM combines metamorphic testing and label information to realize test oracle identification, based on which the ability of MICS in classifying different label combinations is evaluated. To evaluate the effectiveness of ATOM, we have performed experiments on two popular datasets of MICS, VOC and COCO (each with five state-of-the-art DNN models), and one real-world photo tagging application from our industrial partner. The experimental results reveal that the performance of current DNN-based MICSs remains less satisfactory even in recognizing correlations between only two labels, as ATOM triggers a total number of 6,049 such label combination related errors for all MICSs studied. In particular, ATOM reports 587 error-revealing images for the industrial MICS, in which 92% of them are confirmed by the developers.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Safe Online Convex Optimization with Heavy-Tailed Observation Noises 39

Safe Online Convex Optimization with Heavy-Tailed Observatio...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Yang, Yunhao Xue, Bo Hao, Yunzhi Li, Ying Wan, Yuanyu School of Software Technology Zhejiang University Ningbo China Department of Computer Science City University of Hong Kong China Big Graph Center School of Computer and Computing Science Hangzhou City University Hangzhou China State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China Bangsheng Technology Co. Ltd. Hangzhou China Institute of Blockchain and Data Security Hangzhou China

ISBN: (纸本)157735897X

We investigate safe online convex optimization (SOCO), where each decision must satisfy a set of unknown linear constraints. Assuming that the unknown constraints can be observed with a sub-Gaussian noise for each chosen decision, previous studies have established a high-probability regret bound of O(T2/3). However, this assumption may not hold in many practical scenarios. To address this limitation, in this paper, we relax the assumption to allow any noise that admits finite (1+ϵ)-th moments for some ϵ ∈ (0, 1], and propose two algorithms that enjoy an O(Tcϵ) regret bound with high probability, where T is the time horizon and cϵ = (1+ϵ)/(1+2ϵ). The key idea of our two algorithms is to respectively utilize the median-of-means and truncation techniques to achieve accurate estimation under heavy-tailed noises. To the best of our knowledge, these are the first algorithms designed to handle SOCO with heavy-tailed observation noises. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

Generalized Image Embedding for Multi-Domain Image Retrieval

Generalized Image Embedding for Multi-Domain Image Retrieval

引用

International Conference on computer Supported Cooperative Work in Design

作者： Boao Xiao Siyuan Wu Xin He Wanchun Dou State Key Laboratory for Novel Software Technology Nanjing University China School of Computer Science Nanjing University of Posts and Telecommunications China

Image embedding, being a fundamental task in computer vision, plays a crucial role in various downstream tasks such as image retrieval. Widely adopted in e-commerce and social media collaboration, image retrieval benefits greatly from representations learned by the embedding model. However, conventional embedding models are often trained on a single domain, leading to inadequate performance in the multi-domain scenario. To address this challenge, we introduce a generalized image embedding model designed for multi-domain image retrieval. The proposed method employs a contrastively learned Vision Transformer and a carefully crafted training scheme to enhance domain generalization capability. Our theoretical analysis and experimental results, conducted on a large-scale, real-world multi-domain image retrieval dataset, demonstrate the superiority of the proposed method over existing embedding models in terms of both accuracy and domain generalization capability.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：