检索结果-内蒙古大学图书馆

Enhancing EEG Artifact Removal Efficiency by Introducing Dense Skip Connections to IC-U-Net

学校读者我要写书评

暂无评论

Enhancing EEG Artifact Removal Efficiency by Introducing Den...

Annual International Conference of the IEEE engineering in Medicine and Biology Society (EMBC)

作者： Kong-Yi Chang Yung-Chia Huang Chun-Hsiang Chuang Institute of Information Systems and Applications College of Electrical Engineering and Computer Science NTHU Taiwan Research Center for Education and Mind Sciences College of Education NTHU Taiwan Department of Engineering and System Science NTHU Taiwan

Electroencephalographic (EEG) data is considered contaminated with various types of artifacts. Deep learning has been successfully applied to developing EEG artifact removal techniques to increase the signal-to-noise ratio (SNR) and enhance brain-computer interface performance. Recently, our research team has proposed an end-to-end UNet-based EEG artifact removal technique, IC-U-Net, which can reconstruct signals against various artifacts. However, this model suffers from being prone to overfitting with a limited training dataset size and demanding a high computational cost. To address these issues, this study attempted to leverage the architecture of UNet++ to improve the practicability of IC-U-Net by introducing dense skip connections in the encoder-decoder architecture. Results showed that this proposed model obtained superior SNR to the original model with half the number of parameters. Also, this proposed model achieved comparable convergency using a quarter of the training data size.

关键词：

Technical Report: On the Convergence of Gossip Learning in the Presence of Node Inaccessibility

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Tian Cui, Yue Hu, Xueyang Xu, Yecheng Liu, Bo Zhejiang Laboratory Hangzhou China Department of Computer Science and Computer Information Systems Auburn University at Montgomery Montgomery United States Department of Computer Science and Software Engineering Auburn University Auburn United States Institute of Rural Development Zhejiang Academy of Agricultural Sciences Hangzhou China DBAPPSecurity Co. Ltd. Hangzhou China

Gossip learning (GL), as a decentralized alternative to federated learning (FL), is more suitable for resource-constrained wireless networks, such as Flying Ad-Hoc Networks (FANETs) that are formed by unmanned aerial vehicles (UAVs). GL can significantly enhance the efficiency and extend the battery life of UAV networks. Despite the advantages, the performance of GL is strongly affected by data distribution, communication speed, and network connectivity. However, how these factors influence the GL convergence is still unclear. Existing work studied the convergence of GL based on a virtual quantity for the sake of convenience, which failed to reflect the real state of the network when some nodes are inaccessible. In this paper, we formulate and investigate the impact of inaccessible nodes to GL under a dynamic network topology. We first decompose the weight divergence by whether the node is accessible or not. Then, we investigate the GL convergence under the dynamic of node accessibility and theoretically provide how the number of inaccessible nodes, data non-i.i.d.-ness, and duration of inaccessibility affect the convergence. Extensive experiments are carried out in practical settings to comprehensively verify the correctness of our theoretical findings. Copyright © 2024, The Authors. All rights reserved.

关键词： Antennas

Privacy-Aware Joint DNN Model Deployment and Partition Optimization for Delay-Efficient Collaborative Edge Inference

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Cheng, Zhipeng Xia, Xiaoyu Wang, Hong Liwang, Minghui Chen, Ning Fan, Xuwei Wang, Xianbin School of Future Science and Engineering Soochow University Suzhou215006 China School of Computing Technologies RMIT University MelbourneVIC Australia Department of Control Science and Engineering The National Key Laboratory of Autonomous Intelligent Unmanned Systems Tongji University Shanghai201804 China Frontiers Science Center for Intelligent Autonomous Systems Ministry of Education Tongji University Shanghai201804 China Qingdao China College of Computer and Information Sciences Fujian Agriculture and Forestry University Fuzhou350002 China Department of Electrical and Computer Engineering Western University ON Canada

Edge inference (EI) is a key solution to address the growing challenges of delayed response times, limited scalability, and privacy concerns in cloud-based Deep Neural Network (DNN) inference. However, deploying DNN models on resource-constrained edge devices faces more severe challenges, such as model storage limitations, dynamic service requests, and privacy risks. This paper proposes a novel framework for privacy-aware joint DNN model deployment and partition optimization to minimize long-term average inference delay under resource and privacy constraints. Specifically, the problem is formulated as a complex optimization problem considering model deployment, user-server association, and model partition strategies. To handle the NP-hardness and future uncertainties, a Lyapunov-based approach is introduced to transform the long-term optimization into a single-time-slot problem, ensuring system performance. Additionally, a coalition formation game model is proposed for edge server association, and a greedy-based algorithm is developed for model deployment within each coalition to efficiently solve the problem. Extensive simulations show that the proposed algorithms effectively reduce inference delay while satisfying privacy constraints, outperforming baseline approaches in various scenarios. Copyright © 2025, The Authors. All rights reserved.

关键词： Differential privacy

RIEMANNIAN GEOMETRY-BASED EEG APPROACHES: A LITERATURE REVIEW

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Tibermacine, Imad Eddine Russo, Samuele Tibermacine, Ahmed Rabehi, Abdelaziz Nail, Bachir Kadri, Kamel Napoli, Christian Department of Computer Automation and Management Engineering Sapienza University of Rome Rome Italy Department of Psychology Sapienza University of Rome Rome Italy Department of Computer Science University of Biskra Biskra Algeria Faculty of Science and Technology University of Djelfa Djelfa Algeria Polytechnic Military School Algiers Algeria Institute for Systems Analysis and Computer Science Italian National Research Council Italy Department of Computational Intelligence Czestochowa University of Technology Poland

The application of Riemannian geometry in the decoding of brain-computer interfaces (BCIs) has swiftly garnered attention because of its straightforwardness, precision, and resilience, along with its aptitude for transfer learning, which has been demonstrated through significant achievements in global BCI competitions. This paper presents a comprehensive review of recent advancements in the integration of deep learning with Riemannian geometry to enhance EEG signal decoding in BCIs. Our review updates the findings since the last major review in 2017, comparing modern approaches that utilize deep learning to improve the handling of non-Euclidean data structures inherent in EEG signals. We discuss how these approaches not only tackle the traditional challenges of noise sensitivity, non-stationarity, and lengthy calibration times but also introduce novel classification frameworks and signal processing techniques to reduce these limitations significantly. Furthermore, we identify current shortcomings and propose future research directions in manifold learning and riemannian-based classification, focusing on practical implementations and theoretical expansions, such as feature tracking on manifolds, multitask learning, feature extraction, and transfer learning. This review aims to bridge the gap between theoretical research and practical, real-world applications, making sophisticated mathematical approaches accessible and actionable for BCI enhancements. © 2024, CC BY.

关键词： Geometry

BER Improvement of Cooperative Spectrum Sharing of NOMA in 5G Network

学校读者我要写书评

暂无评论

BER Improvement of Cooperative Spectrum Sharing of NOMA in 5...

sciences and Techniques of Automatic Control and computer engineering MI-STA, IEEE International Maghreb Meeting of the Conference on

作者： Mohamed Hassan Rashid A. Saeed Rania A. Mokhtar Khalid Hamid Elmustafa Sayed Ali Nidhal Odeh Manwinder Singh Othman O. Khalifa Shayla Islam Department of Wireless Communication Lovely Professional University Phagwara Punjab India Department of Computer Engineering College of Computers and Information Technology Taif University Taif Saudi Arabia Department of Communication Systems Engineering University of Science & Technology Omdurman Sudan Department of Electronics Engineering Sudan University of Science and Technologu (SUST) Khartoum Sudan School of Electrical Computer and Telecommunications Engineering University of Wollongong Australia Department of Electrical and Computer Engineering International Islamic University Malaysia Gombak Malaysia Libyan Center for Engineering Resarch and Information Technology Bani Waleed Libya Department of Computer Science Institute of Computer Science and Digital Innovation UCSI University Malaysia Kuala Lumpur Malaysia

Non-orthogonal multiple access is one of the best methods for addressing the needs of 5G wireless services (NOMA). The proliferation of mobile devices in recent years has increased the importance of cooperative spectrum sharing and utilization in wireless communication due to increases in the bit error rate (BER) brought on by collisions and interference. In this study, MIMO and Massive MIMO (M-MIMO) in the downlink (DL) NOMA power domain (PD) in conjunction with the Cooperative Cognitive Radio Network are proposed as two new ways for improving and evaluating BER in the 5G network (CCRN). Customers of NOMA compete for available channels on the CCRN in this first strategy. The second approach creates a dedicated channel for NOMA users. Three scenarios with varying distances, power location coefficients, and transmission power are used to evaluate the proposed methods in the MATLAB software program. It is assumed that four users will share a 90 MHz BW using QPSK modulation in all three scenarios. When evaluating the performance of the suggested system under the presumption of frequency-selective Rayleigh fading, concurrent interference cancellation (SIC) and unstable channel conditions are also considered.

关键词：

Graph Auto-Encoder Via Neighborhood Wasserstein Reconstruction

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Tang, Mingyue Yang, Carl Li, Pan Department of Engineering Systems and Environment University of Virginia United States Department of Computer Science Emory University United States Department of Computer Science Purdue University United States

Graph neural networks (GNNs) have drawn significant research attention recently, mostly under the setting of semi-supervised learning. When task-agnostic representations are preferred or supervision is simply unavailable, the auto-encoder framework comes in handy with a natural graph reconstruction objective for unsupervised GNN training. However, existing graph auto-encoders are designed to reconstruct the direct links, so GNNs trained in this way are only optimized towards proximity-oriented graph mining tasks, and will fall short when the topological structures matter. In this work, we revisit the graph encoding process of GNNs which essentially learns to encode the neighborhood information of each node into an embedding vector, and propose a novel graph decoder to reconstruct the entire neighborhood information regarding both proximity and structure via Neighborhood Wasserstein Reconstruction (NWR). Specifically, from the GNN embedding of each node, NWR jointly predicts its node degree and neighbor feature distribution, where the distribution prediction adopts an optimal-transport loss based on the Wasserstein distance. Extensive experiments on both synthetic and real-world network datasets show that the unsupervised node representations learned with NWR have much more advantageous in structure-oriented graph mining tasks, while also achieving competitive performance in proximity-oriented ones. Copyright © 2022, The Authors. All rights reserved.

关键词： Graph neural networks

Trends, Opportunities, and Challenges in Detecting Depressive Disorders Through Mobile Devices: A Review

学校读者我要写书评

暂无评论

Trends, Opportunities, and Challenges in Detecting Depressiv...

computer System, Information Technology, and Electrical engineering (COSITE), International Conference on

作者： Gregorius Natanael Elwirehardja Mahmud Isnan Anzaludin Samsinga Perbangsa Kahlil Muchtar Bens Pardamean Computer Science Department School of Computer Science Bina Nusantara University Jakarta Indonesia Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta Indonesia Information System Department School of Information Systems Bina Nusantara University Jakarta Indonesia Department of Electrical and Computer Engineering Universitas Syiah Kuala Banda Aceh Indonesia Computer Science Department BINUS Graduate Program -Master of Computer Science Bina Nusantara University Jakarta Indonesia

Depressive Disorders (DD) is one of the most prevalent mental disorders in the world that may lead to suicide cases. To prevent the latter, ubiquitous early detection systems may be effective. Recent studies have since researched the development of such systems by exploiting several forms of data, including video, audio, Ecological Momentary Assessments (EMA), and passive sensing data using sensors embedded in mobile devices. To summarize the trends, opportunities, and existing challenges in this field, this study reviewed 15 papers to answer four research questions. EMA was the most popular data to be used in this task, but other approaches, such as using video, audio, and typing behaviors, may be considered due to the subjectivity of EMA. These data were typically recorded using smartphones and analyzed using Machine Learning (ML). However, most of the developed systems had yet to be implemented. Overall, it was concluded that further studies may need to explore usages of more objective data in multimodal approaches as well as consider using Mobile Cloud Computing (MCC) to deploy these systems to provide more effective and efficient diagnoses. Future studies must also take into account the existing challenges of the data and infrastructures, such as the weaknesses of several data types, limitations of mobile devices, as well as the challenges of diagnosis approaches.

关键词：

Uncovering Neural Scaling Laws in Molecular Representation Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Chen, Dingshuo Zhu, Yanqiao Zhang, Jieyu Du, Yuanqi Li, Zhixun Liu, Qiang Wu, Shu Wang, Liang Center for Research on Intelligent Perception and Computing Institute of Automation Chinese Academy of Sciences China Department of Computer Science University of California Los Angeles United States The Paul G. Allen School of Computer Science and Engineering University of Washington United States Department of Computer Science Cornell University United States Department of Systems Engineering and Engineering Management The Chinese University of Hong Kong Hong Kong

Molecular Representation Learning (MRL) has emerged as a powerful tool for drug and materials discovery in a variety of tasks such as virtual screening and inverse design. While there has been a surge of interest in advancing model-centric techniques, the influence of both data quantity and quality on molecular representations is not yet clearly understood within this field. In this paper, we delve into the neural scaling behaviors of MRL from a data-centric viewpoint, examining four key dimensions: (1) data modalities, (2) dataset splitting, (3) the role of pre-training, and (4) model capacity. Our empirical studies confirm a consistent power-law relationship between data volume and MRL performance across these dimensions. Additionally, through detailed analysis, we identify potential avenues for improving learning efficiency. To challenge these scaling laws, we adapt seven popular data pruning strategies to molecular data and benchmark their performance. Our findings underline the importance of data-centric MRL and highlight possible directions for future research. Copyright © 2023, The Authors. All rights reserved.

关键词： Benchmarking

Risk-Averse Self-Scheduling of Storage in Decentralized Markets

学校读者我要写书评

暂无评论

Risk-Averse Self-Scheduling of Storage in Decentralized Mark...

IEEE General Meeting Power& Energy Society

作者： Ogun Yurdakul Farhad Billimoria Energy Systems and Infrastructure Analysis Division Argonne National Laboratory Lemont IL USA Department of Electrical Engineering and Computer Science Technical University of Berlin Berlin Germany Department of Engineering Science University of Oxford Oxford United Kingdom

Storage is expected to be a critical source of firming in low-carbon grids. A common concern raised from ex-post assessments is that storage resources can fail to respond to strong price signals during times of scarcity. While commonly attributed to forecast error or failures in operations, we posit that this behavior can be explained from the perspective of risk-averse scheduling. Using a stochastic self-scheduling framework and real-world data harvested from the Australian National Electricity Market, we demonstrate that risk-averse storage resources tend to have a myopic operational perspective, that is, they typically engage in near-term price arbitrage and chase only few extreme price spikes and troughs, thus remaining idle in several time periods with markedly high and low prices. This has important policy implications given the non-transparency of unit risk aversion and apparent risk in intervention decision-making.

关键词：