检索结果-内蒙古大学图书馆

Dual-Correction Adaptation Network for Noisy Knowledge Transfer

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Wang, Yunyun Zheng, Weiwen Chen, Songcan The Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Computer Science and Engineering Nanjing University of Posts & Telecommunications Nanjing210046 China The MIIT Key Laboratory of Pattern Analysis and Machine Intelligence Computer Science & Technology/AI Nanjing University of Aeronautics & Astronautics Nanjing210023 China

Previous unsupervised domain adaptation (UDA) methods aim to promote target learning via a single-directional knowledge transfer from label-rich source domain to unlabeled target domain, while its reverse adaption from target to source has not jointly been considered yet so far. In fact, in some real teaching practice, a teacher helps students learn while also gets promotion from students to some extent, which inspires us to explore a dual-directional knowledge transfer between domains, and thus propose a Dual-Correction Adaptation Network (DualCAN) in this paper. However, due to the asymmetrical label knowledge across domains, transfer from unlabeled target to labeled source poses a more difficult challenge than the common source-to-target counterpart. First, the target pseudo-labels predicted by source commonly involve noises due to model bias, hence in the reverse adaptation, they may hurt the source performance and bring a negative target-to-source transfer. Secondly, source domain usually contains innate noises, which will inevitably aggravate the target noises, leading to noise amplification across domains. To this end, we further introduce a Noise Identification and Correction (NIC) module to correct and recycle noises in both domains. To our best knowledge, this is the first naive attempt of dual-directional adaptation for noisy UDA, and naturally applicable to noise-free UDA. A theory justification is given to state the rationality of our intuition. Empirical results confirm the effectiveness of DualCAN with remarkable performance gains over state-of-the-arts, particularly for extreme noisy tasks (e.g., ∼+ 15% on Pw→Pr and Pr→Rw of Office-Home). Copyright © 2022, The Authors. All rights reserved.

关键词： Knowledge management

EDSep: An Effective Diffusion-Based Method for Speech Source Separation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Dong, Jinwei Wang, Xinsheng Mao, Qirong School of Computer Science and Communication Engineering Jiangsu University China Jiangsu Engineering Research Center of Big Data Ubiquitous Perception and Intelligent Agriculture Applications China Provincial Key Laboratory of Computational Intelligence and New Technologies in Low-Altitude Digital Agriculture Zhenjiang China Audio Speech and Language Processing Group School of Computer Science Northwestern Polytechnical University Xi’an China

关键词： Stochastic systems

Unveiling Optimal SDG Pathways: An Innovative Approach Leveraging Graph Pruning and Intent Graph for Effective Recommendations

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yu, Zhihang Wang, Shu Zhu, Yunqiang Yuan, Wen Dai, Xiaoliang Zou, Zhiqiang College of Computer Nanjing University of Posts and Telecommunications Nanjing210023 China State Key Laboratory of Resources and Environmental Information System Institute of Geographic Sciences and Natural Resources Research Chinese Academy of Sciences Beijing China Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing210023 China University of Chinese Academy of Sciences Beijing China

The recommendation of appropriate development pathways, also known as ecological civilization patterns for achieving Sustainable Development Goals (namely, sustainable development patterns), are of utmost importance for promoting ecological, economic, social, and resource sustainability in a specific region. To achieve this, the recommendation process must carefully consider the region's natural, environmental, resource, and economic characteristics. However, current recommendation algorithms in the field of computer science fall short in adequately addressing the spatial heterogeneity related to environment and sparsity of regional historical interaction data, which limits their effectiveness in recommending sustainable development patterns. To overcome these challenges, this paper proposes a method called User Graph after Pruning and Intent Graph (UGPIG). Firstly, we utilize the high-density linking capability of the pruned User Graph to address the issue of spatial heterogeneity neglect in recommendation algorithms. Secondly, we construct an Intent Graph by incorporating the intent network, which captures the preferences for attributes including environmental elements of target regions. This approach effectively alleviates the problem of sparse historical interaction data in the region. Through extensive experiments, we demonstrate that UGPIG outperforms state-of-the-art recommendation algorithms like KGCN, KGAT, and KGIN in sustainable development pattern recommendations, with a maximum improvement of 9.61% in Top-3 recommendation performance. Copyright © 2023, The Authors. All rights reserved.

关键词： Sustainable development

Online Attentive Kernel-Based Temporal Difference Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Yang, Guang Chen, Xingguo Yang, Shangdong Wang, Huihui Dong, Shaokang Gao, Yang The the Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications National Engineering Laboratory for Agri-Product Quality Traceability Beijing Technology and Business University China The State Key Laboratory for Novel Software Technology Nanjing University China The PCA Lab Key Lab of Intelligent Perception and Systems for High-Dimensional Information Ministry of Education Jiangsu Key Lab of Image and Video Understanding for Social Security School of Computer Science and Engineering Nanjing University of Science and Technology China

With rising uncertainty in the real world, online Reinforcement Learning (RL) has been receiving increasing attention due to its fast learning capability and improving data efficiency. However, online RL often suffers from complex Value Function Approximation (VFA) and catastrophic interference, creating difficulty for the deep neural network to be applied to an online RL algorithm in a fully online setting. Therefore, a simpler and more adaptive approach is introduced to evaluate value function with the kernel-based model. Sparse representations are superior at handling interference, indicating that competitive sparse representations should be learnable, non-prior, non-truncated and explicit when compared with current sparse representation methods. Moreover, in learning sparse representations, attention mechanisms are utilized to represent the degree of sparsification, and a smooth attentive function is introduced into the kernel-based VFA. In this paper, we propose an Online Attentive Kernel-Based Temporal Difference (OAKTD) algorithm using two-timescale optimization and provide convergence analysis of our proposed algorithm. Experimental evaluations showed that OAKTD outperformed several Online Kernel-based Temporal Difference (OKTD) learning algorithms in addition to the Temporal Difference (TD) learning algorithm with Tile Coding on public Mountain Car, Acrobot, CartPole and Puddle World tasks. Copyright © 2022, The Authors. All rights reserved.

关键词： Learning algorithms

ReconBoost: Boosting Can Achieve Modality Reconcilement

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Hua, Cong Xu, Qianqian Bao, Shilong Yang, Zhiyong Huang, Qingming Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China School of Cyber Security University of Chinese Academy of Sciences Beijing China Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

This paper explores a novel multi-modal alternating learning paradigm pursuing a reconciliation between the exploitation of uni-modal features and the exploration of cross-modal interactions. This is motivated by the fact that current paradigms of multi-modal learning tend to explore multi-modal features simultaneously. The resulting gradient prohibits further exploitation of the features in the weak modality, leading to modality competition, where the dominant modality overpowers the learning process. To address this issue, we study the modality-alternating learning paradigm to achieve reconcilement. Specifically, we propose a new method called ReconBoost to update a fixed modality each time. Herein, the learning objective is dynamically adjusted with a reconcilement regularization against competition with the historical models. By choosing a KL-based reconcilement, we show that the proposed method resembles Friedman's Gradient-Boosting (GB) algorithm, where the updated learner can correct errors made by others and help enhance the overall performance. The major difference with the classic GB is that we only preserve the newest model for each modality to avoid overfitting caused by ensembling strong learners. Furthermore, we propose a memory consolidation scheme and a global rectification scheme to make this strategy more effective. Experiments over six multi-modal benchmarks speak to the efficacy of the method. We release the code at https://***/huacong/ReconBoost. Copyright © 2024, The Authors. All rights reserved.

关键词： Machine learning

QoI-aware incentive for multimedia crowdsensing enabled learning system

学校读者我要写书评

暂无评论

QoI-aware incentive for multimedia crowdsensing enabled lear...

作者： Gu, Yiren Shen, Hang Bai, Guangwei Wang, Tianjing Liu, Xuejun College of Computer Science and Technology Nanjing Tech University Nanjing211816 China Jiangsu Key Laboratory of Big Data Security and Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing210023 China

While much research has been devoted to algorithm improvement of the machine learning model for multimedia applications, relatively little research has focused on the acquisition of massive multimedia datasets with strict data demands for model training. In this paper, we propose a Quality-of-Information (QoI) aware incentive mechanism in multimedia crowdsensing, with the objective of promoting the growth of an initial training model. We begin with a reverse auction incentive model to maximize social welfare while meeting the requirements in quality, timeliness, correlation, and coverage. Then, we discuss how to achieve the optimal social welfare in the presence of an NP-hard winner determination problem. Lastly, we design an effective incentive mechanism to solve the auction problem, which is shown to be truthful, individually rational and computationally efficient. Our evaluation study is carried out using a real multimedia dataset. Extensive simulation results demonstrate that the proposed incentive mechanism produces close-to-optimal social welfare noticeably, while accompanied by accelerating the growth of the machine learning model with a high-QoI dataset. © 2019, Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Machine learning

Minimum Time Extrema Estimation for Large-Scale Radio-Frequency Identification Systems

学校读者我要写书评

暂无评论

Journal of Computer Science & Technology 2020年第5期35卷 1099-1114页

作者： Xiao-Jun Zhu Li-Jie Xu Xiao-Bing Wu Bing Chen College of Computer Science and Technology Nanjing University of Aeronautics and AstronauticsNanjing 211106China Jiangsu Key Laboratory of Dig Data Security and Intelligent Processing Nanjing University of Posts and TelecommunicationsNanjing 210049China Wireless Research Centre University of CanterburyChristchurch 8041New Zealand

We consider the extrema estimation problem in large-scale radio-frequency identification(RFID)systems,where there are thousands of tags and each tag contains a finite *** objective is to design an extrema estimation protocol with the minimum execution *** the standard binary search protocol wastes much time due to inter-frame overhead,we propose a parameterized protocol and treat the number of slots in a frame as an unknown *** formulate the problem and show how to find the best parameter to minimize the worst-case execution ***,we propose two rules to further reduce the execution *** first is to find and remove redundant *** second is to concatenate a frame from minimum value estimation with a frame from maximum value estimation to reduce the total number of *** show that,in a typical scenario,the proposed protocol reduces execution time by 79%compared with the standard binary search protocol.

关键词： radio-frequency identification(RFID)system maximum value estimation minimum value estimation time efficient protocol

ReconBoost: boosting can achieve modality reconcilement 24

学校读者我要写书评

暂无评论

ReconBoost: boosting can achieve modality reconcilement

Proceedings of the 41st International Conference on Machine Learning

作者： Cong Hua Qianqian Xu Shilong Bao Zhiyong Yang Qingming Huang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China Institute of Information Engineering Chinese Academy of Sciences Beijing China and School of Cyber Security University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China School of Computer Science and Technology University of Chinese Academy of Sciences Beijing China and Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing China and Key Laboratory of Big Data Mining and Knowledge Management Chinese Academy of Sciences Beijing China

关键词：

Network data Classification Mechanism for Intrusion Detection System

学校读者我要写书评

暂无评论

Network Data Classification Mechanism for Intrusion Detectio...

International Conference on Computer Supported Cooperative Work in Design

作者： Shuai Jiang Xiaolong Xu Jiangsu Key Laboratory of Big Data Security & Intelligent Processing Nanjing University of Posts and Telecommunications Nanjing China School of Computer Science Nanjing University of Posts and Telecommunications Nanjing China

Intrusion detection system (IDS), as a network security device, monitors network data in real time and responds actively when it detects suspicious transmissions. However, suffered from the large amount of redundancy and high correlation of network data, the traditional IDS have defects in low detection rate and high computational overhead. In this paper, we propose a network data classification mechanism (CPEL). data preprocessing of network traffic data is first performed using correlation-based feature selection (CFS) and principal component analysis (PCA). CFS selects the best features in the data, PCA to dimensionality reduction and denoising. Then, we use multiple classifiers to perform anomaly detection, and select the best three classifiers according to the classification effect. On this basis, we use the majority vote as the ensemble learning (EL) method of combination rules to further improve the performance of the classifier. Experimental results on the benchmark NSL-KDD and WSN-DS datasets indicate that this mechanism outperforms a single algorithm in accuracy and detection rate. In the meantime, computational overhead is remarkably reduced.

关键词： Dimensionality reduction Wireless sensor networks Redundancy Intrusion detection Telecommunication traffic Feature extraction Real-time systems