检索结果-内蒙古大学图书馆

Towards Few-shot Inductive Link Prediction on Knowledge Graphs: A Relational Anonymous Walk-guided Neural Process Approach

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhao, Zicheng Luo, Linhao Pan, Shirui Nguyen, Quoc Viet Hung Gong, Chen School of Computer Science and Engineering Nanjing University of Science and Technology China Jiangsu Key Laboratory of Image and Video Understanding for Social Security China Department of Data Science and AI Monash University Australia School of Information and Communication Technology Griffith University Australia Key Laboratory of Intelligent Perception and Systems for High-Dimensional Information of Ministry of Education

Few-shot inductive link prediction on knowledge graphs (KGs) aims to predict missing links for unseen entities with few-shot links observed. Previous methods are limited to transductive scenarios, where entities exist in the knowledge graphs, so they are unable to handle unseen entities. Therefore, recent inductive methods utilize the sub-graphs around unseen entities to obtain the semantics and predict links inductively. However, in the few-shot setting, the sub-graphs are often sparse and cannot provide meaningful inductive patterns. In this paper, we propose a novel relational anonymous walk-guided neural process for few-shot inductive link prediction on knowledge graphs, denoted as RawNP. Specifically, we develop a neural process-based method to model a flexible distribution over link prediction functions. This enables the model to quickly adapt to new entities and estimate the uncertainty when making predictions. To capture general inductive patterns, we present a relational anonymous walk to extract a series of relational motifs from few-shot observations. These motifs reveal the distinctive semantic patterns on KGs that support inductive predictions. Extensive experiments on typical benchmark datasets demonstrate that our model derives new state-of-the-art performance. Copyright © 2023, The Authors. All rights reserved.

关键词： Forecasting

Retraction Note: Enabling secure and efficient industry 4.0 transformation through trust-authorized anomaly detection in cloud environments with a hybrid AI approach

学校读者我要写书评

暂无评论

Optical and Quantum Electronics 2024年第10期56卷 1-2页

作者： Prakash, N. Vignesh, J. Ashwin, M. Ramadass, Sudhir Veeranjaneyulu, N. V. Athawale, Shashikant Ravuri, Ananda Subramanian, Balambigai Department of Information Technology B.S Abdur Rahman Crescent Institute of Science and Technology Vandalur India Department of ISE Jain University Bengaluru India Deparment of Artificial Intelligence and Data Science Koneru Lakshmaiah Education Foundation Vaddeswaram India Sterck Systems Chennai India Department of Information Technology VFSTR Deemed to Be University Vadlamudi Guntur District India Department of Computer Engineering AISSMS COE Pune India Intel Corporation Hillsboro USA Department of ECE Kongu Engineering College Perundurai India

SWIPT-Enabled MISO Ad Hoc Network Underlay RSMA-based Cellular Network with IRS

学校读者我要写书评

暂无评论

SWIPT-Enabled MISO Ad Hoc Network Underlay RSMA-based Cellul...

IEEE Conference on Vehicular Technology (VTC)

作者： Nguyen Thi Thanh Van Nguyen Cong Luong Feng Shaohan Shimin Gong Dusit Niyato Dong In Kim Faculty of Computer Science PHENIKAA University Hanoi Vietnam School of Information and Electronic Engineering Zhejiang Gongshang University Hangzhou China School of Intelligent Systems Engineering Sun Yat-sen University Guangzhou China College of Computing and Data Sciene Nanyang Technological University Singapore Department of Electrical and Computer Engineering Sungkyunkwan University Suwon South Korea

ISBN: (数字)9798350387414

ISBN: (纸本)9798350387421

In this paper, we propose a simultaneous wire-less information and power transfer (SWIPT)-enabled Ad hoc network underlay rate-splitting multiple access (RSMA)-based system with intelligent reflecting surface (IRS). Therein, a base station (BS) in a primary network uses RSMA to serve primary users (PUs), and secondary user (SU) pairs constitute an Ad hoc network sharing the spectrum with the primary network. The power splitting (PS)-based SWIPT protocol is used in the Ad hoc network that allow the SU receivers to decode the information and harvest energy simultaneously. An IRS is deployed to further enhance the system performance. We formulate optimization problems that optimize the common data rate allocation and beamformers associated with the common and private messages of RSMA, the beamformers and the PS factor in the Ad hoc network, and reflection coefficients of the IRS to maximize the minimum rate of the PUs while satisfying the requirements of harvested energy and data rate of the SU pairs. The optimization problems are non-convex and challenging to be solved. We propose a low complexity algorithm based on alternating descent techniques. Numerical results demonstrate the effectiveness and improvement of the proposed framework compared with the framework based on existing multiple access schemes, i.e., non-orthogonal multiple access (NOMA).

关键词： Cellular networks NOMA Vehicular and wireless technologies System performance Simulation Throughput Ad hoc networks

Noise-Robust and Resource-Efficient ADMM-based Federated Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lari, Ehsan Arablouei, Reza Gogineni, Vinay Chakravarthi Werner, Stefan The Department of Electronic Systems Norwegian University of Science and Technology Trondheim7491 Norway The Department of Information and Communications Engineering Aalto University 00076 Finland The Commonwealth Scientific and Industrial Research Organisation PullenvaleQLD4069 Australia The SDU Applied AI and Data Science The Maersk Mc-Kinney Moller Institute Faculty of Engineering University of Southern Denmark Odense5230 Denmark

Federated learning (FL) leverages client-server communications to train global models on decentralized data. However, communication noise or errors can impair model accuracy. To address this problem, we propose a novel FL algorithm that enhances robustness against communication noise while also reducing communication load. We derive the proposed algorithm through solving the weighted least-squares (WLS) regression problem as an illustrative example. We first frame WLS regression as a distributed convex optimization problem over a federated network employing random scheduling for improved communication efficiency. We then apply the alternating direction method of multipliers (ADMM) to iteratively solve this problem. To counteract the detrimental effects of cumulative communication noise, we introduce a key modification by eliminating the dual variable and implementing a new local model update at each participating client. This subtle yet effective change results in using a single noisy global model update at each client instead of two, improving robustness against additive communication noise. Furthermore, we incorporate another modification enabling clients to continue local updates even when not selected by the server, leading to substantial performance improvements. Our theoretical analysis confirms the convergence of our algorithm in both mean and the mean-square senses, even when the server communicates with a random subset of clients over noisy links at each iteration. Numerical results validate the effectiveness of our proposed algorithm and corroborate our theoretical findings. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

Exploiting NOMA Transmissions in Multi-UAV-assisted Wireless Networks: From Aerial-RIS to Mode-switching UAVs

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhao, Songhan Gong, Shimin Gu, Bo Li, Lanhua Lyu, Bin Hoang, Dinh Thai Yi, Changyan The School of Intelligent Systems Engineering Shenzhen Campus of Sun Yat-sen University China The Guangdong Provincial Key Laboratory of Fire Science and Intelligent Emergency Technology Guangzhou510006 China The School of Communications and Information Engineering Nanjing University of Posts and Telecommunications China The School of Electrical and Data Engineering University of Technology Sydney Australia The College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics China

In this paper, we consider an aerial reconfigurable intelligent surface (ARIS)-assisted wireless network, where multiple unmanned aerial vehicles (UAVs) collect data from ground users (GUs) by using the non-orthogonal multiple access (NOMA) method. The ARIS provides enhanced channel controllability to improve the NOMA transmissions and reduce the co-channel interference among UAVs. We also propose a novel dual-mode switching scheme, where each UAV equipped with both an ARIS and a radio frequency (RF) transceiver can adaptively perform passive reflection or active transmission. We aim to maximize the overall network throughput by jointly optimizing the UAVs’ trajectory planning and operating modes, the ARIS’s passive beamforming, and the GUs’ transmission control strategies. We propose an optimization-driven hierarchical deep reinforcement learning (O-HDRL) method to decompose it into a series of subproblems. Specifically, the multi-agent deep deterministic policy gradient (MADDPG) adjusts the UAVs’ trajectory planning and mode switching strategies, while the passive beamforming and transmission control strategies are tackled by the optimization methods. Numerical results reveal that the O-HDRL efficiently improves the learning stability and reward performance compared to the benchmark methods. Meanwhile, the dual-mode switching scheme is verified to achieve a higher throughput performance compared to the fixed ARIS scheme. Copyright © 2024, The Authors. All rights reserved.

关键词： Unmanned aerial vehicles (UAV)

Resilience in Online Federated Learning: Mitigating Model-Poisoning Attacks via Partial Sharing

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lari, Ehsan Arablouei, Reza Gogineni, Vinay Chakravarthi Werner, Stefan The Department of Electronic Systems Norwegian University of Science and Technology Trondheim7491 Norway Department of Information and Communications Engineering Aalto University 00076 Finland The Commonwealth Scientific and Industrial Research Organisation PullenvaleQLD4069 Australia The SDU Applied AI and Data Science The Maersk Mc-Kinney Moller Institute Faculty of Engineering University of Southern Denmark Odense5230 Denmark

Federated learning (FL) allows training machine learning models on distributed data without compromising privacy. However, FL is vulnerable to model-poisoning attacks where malicious clients tamper with their local models to manipulate the global model. In this work, we investigate the resilience of the partial-sharing online FL (PSO-Fed) algorithm against such attacks. PSO-Fed reduces communication overhead by allowing clients to share only a fraction of their model updates with the server. We demonstrate that this partial sharing mechanism has the added advantage of enhancing PSO-Fed’s robustness to model-poisoning attacks. Through theoretical analysis, we show that PSO-Fed maintains convergence even under Byzantine attacks, where malicious clients inject noise into their updates. Furthermore, we derive a formula for PSO-Fed’s mean square error, considering factors like stepsize, attack probability, and the number of malicious clients. Interestingly, we find a non-trivial optimal stepsize that maximizes PSO-Fed’s resistance to these attacks. Extensive numerical experiments confirm our theoretical findings and showcase PSO-Fed’s superior performance against model-poisoning attacks compared to other leading FL algorithms. Copyright © 2024, The Authors. All rights reserved.

关键词： Federated learning

A Method for Predicting Mechanical Properties of Steel Based on Graph Convolutional Networks

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Ban, Yunqi Yang, Yang Wang, Xianpeng Wu, Zhiyuan National Frontiers Science Center for Industrial Intelligence and Systems Optimization Northeastern University Shenyang110819 China Key Laboratory of Data Analytics and Optimization for Smart Industry Northeastern University Ministry of Education Shenyang110819 China Liaoning Engineering Laboratory of Data Analytics and Optimization for Smart Industry Shenyang China College of Electronic and Information Engineering Tongji University Shanghai201804 China Northeastern University China

In steel materials, several mechanical properties are interrelated and there is an inherent topology structure among these mechanical properties. Based on this motivation, a multitask neural network model based graph convolutional network is proposed to predict mechanical properties. First, a directed graph is built over the mechanical properties, with each mechanical property representing a node on the graph. Then this graph convolutional network, which represents the information propagation among mechanical properties, is combined with a multitask neural network and trained together. By comparing with K-NearestNeighbor (KNN), support vector regression (SVR), random forest (RF), Extreme Gradient Boosting (XGBoost), Cross-Stitch Network, Multi-gate Mixture-of-Experts (MMOE) and multitask neural (MTN) algorithms, the proposed algorithm is optimal in all of the metrics root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2) that indicate the predictive performance on the prediction of tensile strength (TS), yield strength (YS) and elongation (EL). Specifically, R2 is 0.9797 in tensile strength, R2 is 0.9747 in yield strength and R2 is 0.8316 in elongation. In addition, a deep representation of the various influences on mechanical properties is obtained through the parameters of the model with validity and interpretability. Material insights are extracted from this deep representation, including the comprehensive interaction of processing and chemical composition on the mechanical properties as well as the mechanical properties with each other. Further, the prediction accuracy can be improved by the combination of different structures of neural networks with graph convolutional network to prove the effectiveness of graph convolutional network. © 2024, The Authors. All rights reserved.

关键词： Tensile strength

Multilayer perceptron ensembles in a truly sparse training context

学校读者我要写书评

暂无评论

Neural Computing and Applications 2025年 1-20页

作者： van der Wal, Peter R. D. Strisciuglio, Nicola Azzopardi, George Mocanu, Decebal Constatin Information Systems Group Bernoulli Institute for Mathematics Computer Science and Artificial Intelligence University of Groningen Nijenborgh 9 Groningen9747 AG Netherlands Data Management and Biometrics Group Faculty of Electrical Engineering Mathematics and Computer Science University of Twente Hallenweg 19 Enschede7522 NH Netherlands Department of Computer Science Faculty of Science Technology and Medicine University of Luxembourg Maison du Nombre 6 avenue de la Fonte Esch4364 Luxembourg

Ensemble learning for artificial neural networks (ANNs) is an effective method to enhance predictive performance. However, ANNs are computationally and memory intensive, and naively training multiple networks can lead to excessive training times and costs. An effective tool for improving ensemble efficiency is introducing topological sparsity. Even though several implementations of efficient ensembles have been proposed, none of them can provide actual benefits in terms of computational overhead as the sparsity is simulated using binary masks. In this paper, we address this issue by introducing a Truly Sparse Ensemble without binary masks and directly incorporate native sparsity. We also propose two algorithms for initializing new subnetworks within the ensemble, leveraging this native topological sparsity to enhance subnetwork diversity. We demonstrate the performance of the resulting models at high levels of sparsity on several datasets in terms of classification accuracy, floating point operations (FLOPs), and actual running time. The proposed methods outperform all baseline dense and truly sparse models on tabular data, successfully diversify the training trajectory of the subnetworks, and increase the topological distance between subnetworks after re-initialization. © The Author(s) 2025.

关键词： Digital arithmetic

Overview and Assessment Approaches of Energy Consumption and GHG Emissions Through the Development and Application of Artificial Intelligence - An Example from the Agricultural Sector

学校读者我要写书评

暂无评论

Overview and Assessment Approaches of Energy Consumption and...

IEEE International Workshop on Metrology for Agriculture and Forestry (MetroAgriFor)

作者： Stefan Rilling Jens Henningsen Katharina Milde Daniel Martini Lorenz Wickert Patricia Kelbert Adaptive Reflective Teams Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS Sankt Augustin Germany Data Science Fraunhofer Institute for Experimental Software Engineering IESE Kaiserslautern Germany Team Digital Technologies Advisory Board for Technology and Construction in Agriculture KTBL Darmstadt Germany

ISBN: (数字)9798350355444

ISBN: (纸本)9798350355451

This article looks at the sustainability of various artificial intelligence (AI) methods. In recent months, AI methods have received a great deal of public attention and the benefits for sustainability have been emphasised. However, the ecological impacts of AI methods must also be considered in this debate. This article presents an overview of various approaches to sustainability assessment and addresses issues relating to energy consumption and the associated Carbon dioxide (CO2) emissions, and a Life Cycle Assessment (LCA) calculation for a concrete agricultural AI-use case.

关键词： Life cycle assessment Energy consumption Conferences Forestry Carbon dioxide Metrology Agriculture Artificial intelligence Sustainable development Carbon emissions