检索结果-内蒙古大学图书馆

41st International Conference on Machine Learning, ICML 2024

作者： Zhang, Zixin Qi, Fan Xu, Changsheng School of Computer Science and Engineering Tianjin University of Technology Tianjin China Institute of Automation Chinese Academy of Sciences Beijing China

The remarkable generalization of large-scale models has recently gained significant attention in multimodal ***, deploying heterogeneous large-scale models with different modalities under Federated Learning (FL) to protect data privacy imposes tremendous challenges on clients' limited computation and *** this work, we propose M2FEDSA to address the above *** realize modularized decomposition of large-scale models via Split Learning (SL) and only retain privacy-sensitive modules on clients, alleviating storage *** freezing large-scale models and introducing two specialized lightweight adapters, the models can better focus on task-specific knowledge and enhance modality-specific knowledge, improving the model's adaptability to different tasks while balancing *** addition, M2FEDSA further improves performance by transferring multimodal knowledge to unimodal clients at both the feature and decision levels, which leverages the complementarity of different *** experiments on various multimodal classification tasks validate the effectiveness of our proposed *** code is made available publicly at https://***/M2FedSA/M-2FedSA. Copyright 2024 by the author(s)

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

A Quadratic Step-Up DC-DC Converter Based on Switched Capacitor and Coupled Inductor 7

A Quadratic Step-Up DC-DC Converter Based on Switched Capaci...

引用

7th IEEE International Electrical and Energy Conference, CIEEC 2024

作者： Hao, Shicheng Fu, Yu Li, Shouxiang Ren, Zepei School of Automation Beijing Institute of Technology Beijing China School of Computer and Communication Engineering Northeastern University at Qinhuangdao Qinhuangdao China

ISBN: (纸本)9798350359558

This paper proposes a cascaded high gain DC-DC converter based on the switched capacitor and coupled inductor with merits of high voltage-boost capability, low voltage stress and only one active switch. The combination of the coupled inductor and switched capacitor cell not only improves the voltage gain, but also decreases voltage stress on semiconductor devices. The quadratic structure makes the converter to have a wide voltage-gain range. Theoretical analysis of steady-state operation, voltage-gain derivation and component stress is conducted, and an experimental prototype was built to verify the theoretical analysis and feasibility of the converter. © 2024 IEEE.

关键词： Semiconductor devices

来源：评论

学校读者我要写书评

暂无评论

Region-Based Optimization in Continual Learning for Audio Deepfake Detection 39

Region-Based Optimization in Continual Learning for Audio De...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Chen, Yujie Yi, Jiangyan Fan, Cunhang Tao, Jianhua Ren, Yong Zeng, Siding Zhang, Chu Yuan Yan, Xinrui Gu, Hao Xue, Jun Wang, Chenglong Lv, Zhao Zhang, Xiaohui School of Computer Science and Technology Anhui University China Institute of Automation Chinese Academy of Sciences China Department of Automation Tsinghua University China Beijing National Research Center for lnformation Science and Technology Tsinghua University China

ISBN: (纸本)157735897X

Rapid advancements in speech synthesis and voice conversion bring convenience but also new security risks, creating an urgent need for effective audio deepfake detection. Although current models perform well, their effectiveness diminishes when confronted with the diverse and evolving nature of real-world deepfakes. To address this issue, we propose a continual learning method named Region-Based Optimization (RegO) for audio deepfake detection. Specifically, we use the Fisher information matrix to measure important neuron regions for real and fake audio detection, dividing them into four regions. First, we directly fine-tune the less important regions to quickly adapt to new tasks. Next, we apply gradient optimization in parallel for regions important only to real audio detection, and in orthogonal directions for regions important only to fake audio detection. For regions that are important to both, we use sample proportion-based adaptive gradient optimization. This region-adaptive optimization ensures an appropriate trade-off between memory stability and learning plasticity. Additionally, to address the increase of redundant neurons from old tasks, we further introduce the Ebbinghaus forgetting mechanism to release them, thereby promoting the model’s ability to learn more generalized discriminative features. Experimental results show our method achieves a 21.3 percent improvement in EER over the state-of-the-art continual learning approach RWM for audio deepfake detection. Moreover, the effectiveness of RegO extends beyond the audio deepfake detection domain, showing potential significance in other tasks, such as image recognition. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

From News to Summaries: Building a Hungarian Corpus for Extractive and Abstractive Summarization 30

From News to Summaries: Building a Hungarian Corpus for Extr...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Barta, Botond Lakatos, Dorina Nagy, Attila Nyist, Milán Konor Ács, Judit HUN-REN Institute for Computer Science and Control Hungary Department of Automation and Applied Informatics Budapest University of Technology and Economics Hungary

ISBN: (纸本)9782493814104

Training summarization models requires substantial amounts of training data. However for less resourceful languages like Hungarian, openly available models and datasets are notably scarce. To address this gap our paper introduces HunSum-2 an open-source Hungarian corpus suitable for training abstractive and extractive summarization models. The dataset is assembled from segments of the Common Crawl corpus undergoing thorough cleaning, preprocessing and deduplication. In addition to abstractive summarization we generate sentence-level labels for extractive summarization using sentence similarity. We train baseline models for both extractive and abstractive summarization using the collected dataset. To demonstrate the effectiveness of the trained models, we perform both quantitative and qualitative evaluation. Our dataset, models and code are publicly available, encouraging replication, further research, and real-world applications across various domains. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： abstractive summarization extractive summarization Hungarian

来源：评论

学校读者我要写书评

暂无评论

Dynamic Event-Triggered Consensus Control for Input Constrained Multi-Agent Systems With a Designable Minimum Inter-Event Time

引用

IEEE/CAA Journal of Automatica Sinica 2024年第3期11卷 649-660页

作者： Meilin Li Yue Long Tieshan Li Hongjing Liang C.L.Philip Chen the School of Automation Engineering University of Electronic Science and Technology of ChinaChengdu 611731China IEEE the Yangtze Delta Region Institute University of Electronic Science and Technology of ChinaHuzhou 313000 the Laboratory of Electromagnetic Space Cognition and Intelligent Control Beijing 100089China the School of Computer Science and Engineering South China University of TechnologyGuangzhou 510006

This paper investigates the consensus control of multi-agent systems(MASs) with constrained input using the dynamic event-triggered mechanism(ETM).Consider the MASs with small-scale networks where a centralized dynamic ETM with global information of the MASs is first ***,a distributed dynamic ETM which only uses local information is developed for the MASs with large-scale *** is shown that the semi-global consensus of the MASs can be achieved by the designed bounded control protocol where the Zeno phenomenon is eliminated by a designable minimum inter-event *** addition,it is easier to find a trade-off between the convergence rate and the minimum inter-event time by an adjustable ***,the results are extended to regional consensus of the MASs with the bounded control *** simulations show the effectiveness of the proposed approach.

关键词： Constrained input designable minimum inter-event time directed communication topology dynamic event-triggered mechanism MASs consensus control

来源：评论

学校读者我要写书评

暂无评论

PANSA: A Framework for Threat Knowledge Graph Construction 3

PANSA: A Framework for Threat Knowledge Graph Construction

引用

3rd International Conference on Cloud Computing, Big Data Application and Software Engineering, CBASE 2024

作者： Chen, TianXiang Zhuang, Yi Xu, Tao Nanjing University of Aeronautics and Astronautics College of Computer Science and Technology Nanjing China Jiangsu Automation Research Institute Lianyungang China

ISBN: (纸本)9798331506582

Cyber intelligence provides new prevention ideas for cyberattacks, but it is difficult to use effectively due to its multi-source heterogeneity and fragmentation. To address this problem, we propose a threat knowledge graph construction framework based on Pretrained Adversarial NER and relationship Subgraph-based Alignment called PANSA. First, the cyber threat domain ontology model is constructed in combination with the existing cyber threat intelligence standards. Secondly, SecBABC threat entity recognition model is designed, which incorporates pre-training and adversarial training modules to extract high-quality threat intelligence data. Then the Multi-Relation GAT for Entity Alignment framework (MRGEA) is designed. MRGEA considers the impact of different relationships on entities, and uses a multi-relation graph attention network to mine deep information. Finally, the cyber threat knowledge graph is visualized. Experimental results show that PANSA effectively integrates security intelligence and outperforms existing methods in knowledge extraction and entity alignment. © 2024 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Simultaneous Learning Intensity and Optical Flow From High-Speed Spike Stream

引用

IEEE Transactions on Circuits and Systems for Video technology 2025年第6期35卷 5126-5139页

作者： Zhu, Lin Yan, Weiquan Chang, Yi Tian, Yonghong Huang, Hua Beijing Institute of Technology School of Computer Science and Technology Beijing100081 China Peng Cheng Laboratory Shenzhen518055 China Huazhong University of Science and Technology School of Artificial Intelligence and Automation Wuhan430074 China Peking University School of Computer Science Beijing100081 China Beijing Normal University School of Artificial Intelligence Beijing100875 China

Bio-inspired vision sensors, which emulate the human retina by recording light intensity as binary spikes, have gained increasing interest in recent years. Among them, the spike camera is capable of perceiving fine textures by simulating a small retinal region called the fovea and producing high temporal resolution (20,000 Hz) spatiotemporal spike streams. To bridge the gap between binary spike streams and human vision in high-speed scenes, reconstructing intensity and optical flow from high temporal resolution spikes is particularly important. In this paper, we present a hybrid SNN-ANN network designed for simultaneous intensity and optical flow learning from spike streams. To adaptively extract spatial and temporal features from continuous spike streams, we propose a spiking neuron module with dense connections that efficiently processes both short-term and long-term spike data, while maintaining low power consumption characteristics. Subsequently, we introduce two decoders for optical flow and intensity estimation that complement each other. A temporal-aware warping module, based on flow features, is specifically designed to align the temporal features of the intensity decoder, thereby reducing motion artifacts. Concurrently, improved intensity features contribute to more accurate flow feature predictions, resulting in a mutually beneficial relationship within our network. To evaluate the effectiveness of our proposed network, we conduct experiments on both simulated and real spike datasets. Our network outperforms existing state-of-the-art spike-based reconstruction and optical flow estimation methods, demonstrating its potential for advancing the field of bio-inspired vision sensors. © 1991-2012 IEEE.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

Hot Off the Press: Benchmarking Derivative-Free Global Optimization Algorithms under Limited Dimensions and Large Evaluation Budgets

Hot Off the Press: Benchmarking Derivative-Free Global Optim...

引用

2024 Genetic and Evolutionary Computation Conference Companion, GECCO 2024 Companion

作者： Stripinis, Linas Kůdela, Jakub Paulaviçius, Remigijus Institute of Data Science and Digital Technologies Vilnius University Vilnius Lithuania Institute of Automation and Computer Science Brno University of Technology Brno Czech Republic

ISBN: (纸本)9798400704956

This Hot Off the Press paper provides a brief summary of our recent work "Benchmarking Derivative-Free Global Optimization Algorithms under Limited Dimensions and Large Evaluation Budgets"published in IEEE Transactions on Evolutionary Computation [5]. In the paper, we performed a comprehensive computational comparison between stochastic and deterministic global optimization algorithms with twenty-five representative state-of-the-art methods selected from both classes. The experiments were set up with up to twenty dimensions and relatively large evaluation budgets (105 X n). Benchmarking was carried out in a significantly expanded version of the DIRECTGOLib v2.0 library, which included ten distinct collections of primarily continuous test functions. The evaluation of the methods focused on various aspects, such as solution quality, time complexity, and function evaluation usage. The rankings were determined using statistical tests and performance profiles. Our findings suggest that while state-of-the-art deterministic methods could find reasonable solutions with comparatively fewer function evaluations, most stochastic algorithms require more extensive evaluation budgets to deliver comparable results. However, the performance of stochastic algorithms excelled in more complex and higher-dimensional problems. These research findings offer valuable insights for practitioners and researchers, enabling them to tackle diverse optimization problems effectively. © 2024 Copyright held by the owner/author(s).

关键词： Budget control

来源：评论

学校读者我要写书评

暂无评论

Enhancing Bird Migration Studies: Detecting Birdsong in Audio Files Using Convolutional Neural Networks

Enhancing Bird Migration Studies: Detecting Birdsong in Audi...

引用

作者： Honsor, Oksana Gonsor, Yuriy Institute of Computer Technologies Automation and Metrology Lviv Polytechnic National University Lviv79013 Ukraine Institute of Computer Science and Information Technology Lviv Polytechnic National University Lviv79013 Ukraine

Monitoring climate threats and global environmental changes is a crucial issue worldwide. To achieve this, it is important to continuously study these changes using non-standard approaches. One such approach is studying bird migration. It is crucial for analyzing environmental factors that cover large areas of land, such as temperature, air pollution, and water pollution. This article presents a machine-learning model that accurately identifies birdsong in audio files to study bird migration in specific areas. This model uses a convolutional neural network to classify audio files based on images of their spectrograms. The study involves collecting and analyzing audio files that can be used to identify characteristics according to which the sound in the file will be identified as bird sound. The study investigated the effectiveness of using STFT and Mel-spectrograms in a convolutional neural network to classify files based on the presence of bird sounds. The model trained on Mel spectrograms achieved an accuracy of 80%, which is 17% higher than the model trained on STFT spectrograms. This accuracy value is completely sufficient to claim the presence of bird sounds in the audio file. Particular attention is given to the efficiency and accuracy of the CNN model because these indicators enable the comparison and selection of the best classifier for a given file type and model. Although the classification accuracy of the chosen method was lower than expected when building the classification model, the model can predict an audio file in less than 0,5 s, which is a significant achievement. The obtained results may aid in the development of systems for monitoring bird migration across vast territories. To enhance the technique, multiple sets of audio files can be utilized to classify the presence of birds in the environment, improving the accuracy of the method and model. However, the simultaneous complex processing of numerous audio files may reduce processing speed. This

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on a combinatorial control method for coaxial rotor aircraft based on sliding mode

引用

Defence technology（防务技术） 2022年第2期18卷 280-292页

作者： Yi-ran Wei Hong-bin Deng Zhen-hua Pan Ke-wei Li Han Chen School of Mechatronical Engineering Beijing Institute of TechnologyBeijingChina School of Automation Beijing Institute of TechnologyBeijingChina Beijing Institute of Computer and Electronics Application BeijingChina Hong Kong Polytechnic University Hong KongChina

Aiming at the position and attitude tracking of coaxial rotor aircraft(CRA),this paper proposes a combinatorial control method of sliding mode control(SMC)coupled with proportional-integralderivative control(PIDC).Considering the complete description of flight dynamics,aerodynamics and airflow interference,the dynamical model of CRA is *** dynamical model is simplified according to the actual flight,then the simplified dynamical model is divided into two subsystems:a fully-actuated subsystem and an under-actuated *** controller of the fully-actuated subsystem consists of a SMC controller coupled with a rate bounded PIDC controller,while the controller of the under-actuated subsystem is composed of a SMC *** sliding manifold is defined by combining the position and velocity tracking errors of the state variables for each *** stability theory is used to verify the stability of the sliding mode controller,which ensures that all state trajectories of the system can reach and stay on the sliding mode surface,the uncertainty and external interference of the model are *** and experiment compared with the conventional PIDC are carried out,the results demonstrate the effectiveness and the robustness of the proposed control method of this paper.

关键词： Coaxial rotor aircraft Sliding mode control Position and attitude tracking Simulation Experiment

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：