检索结果-内蒙古大学图书馆

Not All Pixels Are Equal: Learning Pixel Hardness for Semantic Segmentation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Xiao, Xin Zhou, Daiguo Hu, Jiagao Hu, Yi Xu, Yongchao National Engineering Research Center for Multimedia Software Institute of Artificial Intelligence School of Computer Science Hubei Key Laboratory of Multimedia and Network Communication Engineering Wuhan University Wuhan430070 China Xiaomi AI Lab Wuhan430070 China Hubei Medical Devices Quality Supervision and Test Institute Hubei Center for AI Medical Devices Supercomputing Wuhan430070 China

Semantic segmentation has recently witnessed great progress. Despite the impressive overall results, the segmentation performance in some hard areas (e.g., small objects or thin parts) is still not promising. A straightforward solution is hard sample mining, which is widely used in object detection. Yet, most existing hard pixel mining strategies for semantic segmentation often rely on pixel’s loss value, which tends to decrease during training. Intuitively, the pixel hardness for segmentation mainly depends on image structure and is expected to be stable. In this paper, we propose to learn pixel hardness for semantic segmentation, leveraging hardness information contained in global and historical loss values. More precisely, we add a gradient-independent branch for learning a hardness level (HL) map by maximizing hardness-weighted segmentation loss, which is minimized for the segmentation head. This encourages large hardness values in difficult areas, leading to appropriate and stable HL map. Despite its simplicity, the proposed method can be applied to most segmentation methods with no and marginal extra cost during inference and training, respectively. Without bells and whistles, the proposed method achieves consistent/significant improvement (1.37% mIoU on average) over most popular semantic segmentation methods on Cityscapes dataset, and demonstrates good generalization ability across domains. The source codes are available at https://***/Menoly-xin/Hardness-Level-Learning. Copyright © 2023, The Authors. All rights reserved.

关键词： Semantic Segmentation

Research of Offloading Decision and Resource Scheduling in Edge Computing Based on Deep Reinforcement Learning 4th

学校读者我要写书评

暂无评论

Research of Offloading Decision and Resource Scheduling in E...

4th EAI International Conference on Smart Grid and Internet of Things, SGIoT 2020

作者： Zhang, Zhen-Jiang Wu, Tong Li, Zhiyuan Shen, Bo Chen, Naiyue Li, Jian Department of Software Engineering Key Laboratory of Communication and Information Systems Beijing Municipal Commission of Education Beijing Jiaotong University Beijing100044 China Department of Electronic and Information Engineering Key Laboratory of Communication and Information Systems Beijing Municipal Commission of Education Beijing Jiaotong University Beijing100044 China School of Computer and Information Technology Beijing Jiaotong University Beijing Municipal Commission of Education Beijing100044 China

ISBN: (纸本)9783030695132

The increasing scale of the IOT poses challenges to the energy consumption, transmission bandwidth and processing delay of centralized cloud computing data centers. The cloud computing data centers is moving from the center of the network to edge nodes with lower latency, namely, edge computing. Meanwhile, it can meet the needs of users for real-time services. In the field of edge computing, offloading decision and resource scheduling are the hot-spot issues. As for offloading decision and resource scheduling problems of single-cell multi-user partial offloading, the system model is also firstly established from four aspects: network architecture, application type, local computing and offloading computing. Based on the system model, the optimization problem of resource scheduling is modeled, where the solution is hard to be found. Thus, the deep reinforcement learning method based on policy gradient is selected to establish the SPBDDPG algorithm that can solve the problem. Then, in order to solve the practical problems, the SPBDDPG algorithm is set up with the state and action for iteration, as well as the environment for generating new state and feedback reward value. Finally, an appropriate iteration step is written for the edge computing resource scheduling problem by combining with the original deep reinforcement learning algorithm. We also evaluate the proposed approaches by relevant experiments. The complexity and effectiveness of the results are validated. © 2021, ICST Institute for computer Sciences, Social Informatics and Telecommunications engineering.

关键词： Reinforcement learning

Towards Evidence-Based Cybersecurity Assessment of Programmable systems to Ensure the Protection of Critical IT Infrastructure

学校读者我要写书评

暂无评论

Towards Evidence-Based Cybersecurity Assessment of Programma...

IEEE International Workshop on Intelligent Data Acquisition and Advanced Computing systems: Technology and Applications

作者： Oleg Illiashenko Vyacheslav Kharchenko Oleg Odarushchenko Department of computer systems and networks National aerospace university “KhAI” Kharkiv Ukraine Software Engineering & Dependable Computing Laboratory Institute of Information Science and Technologies “Alessandro Faedo”—ISTI CNR Pisa Italy Research and Production Enterprise RPC Radics LLC Kropyvnytskyi Ukraine

The paper is devoted to developing scientific principles, methods, means, and information technology of model-oriented verification and evidence-based assessment using functional safety and cybersecurity cases for programmable systems of critical applications (PSCA). In particular, information and control systems important for the safety of nuclear power plants (NPPs), aerospace systems, railway domains, etc., developed using Field Programmable Gate Arrays (FPGAs) and hardware platforms are reviewed. The goal is to ensure the guaranteed completeness and reliability of its functional safety and cybersecurity assessment by developing and implementing a set of formal and semi-formal methods and tools that consider defects of different nature – physical, design, trojans, and vulnerabilities that can be attacked and lead to a fatal system failure, which results in damaging the critical IT infrastructure. The methods shown are based on integrating algebraic, tabular, graph models, and case assessment methodology. The methods are implemented as appropriate technologies for evidence-based verification and evaluation of PSCA. For formal methods, a prototype of a translator of Very High-Speed Integrated Circuits Hardware Description Language (VHDL) code is developed into an algebra of behaviors. It provides evidence-based verification and a framework and tools to design reports on assessing the cybersecurity and functional safety of programmable systems.

关键词：

MEGCF: Multimodal Entity Graph Collaborative Filtering for Personalized Recommendation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Liu, Kang Xue, Feng Guo, Dan Wu, Le Li, Shujie Hong, Richang Hefei University of Technology School of Computer Science and Information Engineering Key Laboratory of Knowledge Engineering with Big Data Intelligent Interconnected Systems Laboratory of Anhui Province 485 Danxia Road Anhui Province Hefei230601 China Hefei University of Technology School of Software Key Laboratory of Knowledge Engineering with Big Data Intelligent Interconnected Systems Laboratory of Anhui Province 485 Danxia Road Anhui Province Hefei230601 China

In most E-commerce platforms, whether the displayed items trigger the user's interest largely depends on their most eye-catching multimodal content. Consequently, increasing efforts focus on modeling multimodal user preference, and the pressing paradigm is to incorporate complete multimodal deep features of the items into the recommendation module. However, the existing studies ignore the mismatch problem between multimodal feature extraction (MFE) and user interest modeling (UIM). That is, MFE and UIM have different emphases. Specifically, MFE is migrated from and adapted to upstream tasks such as image classification. In addition, it is mainly a content-oriented and non-personalized process, while UIM, with its greater focus on understanding user interaction, is essentially a user-oriented and personalized process. Therefore, the direct incorporation of MFE into UIM for purely user-oriented tasks, tends to introduce a large number of preference-independent multimodal noise and contaminate the embedding representations in UIM. This paper aims at solving the mismatch problem between MFE and UIM, so as to generate high-quality embedding representations and better model multimodal user preferences. Towards this end, we develop a novel model, multimodal entity graph collaborative filtering, short for MEGCF. The UIM of the proposed model captures the semantic correlation between interactions and the features obtained from MFE, thus making a better match between MFE and UIM. More precisely, semantic-rich entities are first extracted from the multimodal data, since they are more relevant to user preferences than other multimodal information. These entities are then integrated into the user-item interaction graph. Afterwards, a symmetric linear Graph Convolution Network (GCN) module is constructed to perform message propagation over the graph, in order to capture both high-order semantic correlation and collaborative filtering signals. Finally, the sentiment information fr

关键词： Sentiment analysis

Feature and Performance Comparison of FaaS Platforms

学校读者我要写书评

暂无评论

Feature and Performance Comparison of FaaS Platforms

IEEE International Conference on software engineering and Service Sciences (ICSESS)

作者： Penghui Ma Peichang Shi Guodong Yi National Key Laboratory of Parallel and Distributed Processing College of Computer Science National University of Defense Technology Changsha China Key Laboratory of Software Engineering for Complex Systems College of Computer Science National University of Defense Technology Changsha China Xiangjiang Lab Changsha China School of Advanced Interdisciplinary Studies Hunan University Of Technology and Business Changsha China

With serverless computing offering more efficient and cost-effective application deployment, the diversity of serverless platforms presents challenges to users, including platform lock-in and costly migration. Moreover, due to the black box nature of function computing, traditional performance benchmarking methods are not applicable, necessitating new studies. This article presents a detailed comparison of six major public cloud function computing platforms and introduces a benchmarking framework for function computing performance. This framework aims to help users make comprehensive comparisons and select the most suitable platform for their specific needs.

关键词：

Modeling Neural Networks Training Process with Markov Decision Process 2

学校读者我要写书评

暂无评论

Modeling Neural Networks Training Process with Markov Decisi...

2nd International Conference on Artificial Intelligence and computer engineering, ICAICE 2021

作者： Bai, Yantao Liu, Wanwei Mao, Xinjun Liang, Zhen National University of Defense Technology Key Laboratory of Software Engineering for Complex Systems Changsha China College of Computer Science National University of Defense Technology Changsha China Institute for Quantum Information National University of Defense Technology State Key Laboratory of High Performance Computing Changsha China

ISBN: (纸本)9781665421867

With the development of computer technology, statistics-based machine learning method has made great break-throughs, and also improved the development of artificial intelligence. Nevertheless, as a very influential model, neural networks are still treated as 'black boxes'. The results of neural networks are extremely sensitive to the training samples, which lead to great challenges to the controllability of the algorithm. With the wide application of machine learning, demand for interpretability and controllability of neural networks algorithms is increasing. As a result, various scholars have tried to explain and verify neural networks algorithms based on formal methods in recent years. In this paper, a method (called MNNTP) is presented to model the training process of neural networks by using a Markov decision process (MDP). Through MNNTP, the neural networks are abstracted into the form of MDP, which makes notable contributions for verifying some mathematical properties of the neural networks. © 2021 IEEE.

关键词： Markov processes

Transformer-based Hand Gesture Recognition via High-Density EMG Signals: From Instantaneous Recognition to Fusion of Motor Unit Spike Trains

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Montazerin, Mansooreh Rahimian, Elahe Naderkhani, Farnoosh Atashzar, S. Farokh Yanushkevich, Svetlana Mohammadi, Arash Department of Electrical and Computer Engineering Concordia University MontrealQC Canada Concordia Institute for Information Systems Engineering Concordia University MontrealQC Canada Departments of Electrical & Computer Engineering and Mechanical & Aerospace Engineering New YorkNY10003 United States New YorkNY10003 United States Department of Electrical and Software Engineering Biometric Technologies Laboratory Schulich School of Engineering University of Calgary CalgaryAB Canada

Designing efficient and labor-saving prosthetic hands requires powerful hand gesture recognition algorithms that can achieve high accuracy with limited complexity and latency. In this context, the paper proposes a compact deep learning framework referred to as the CT-HGR, which employs a vision transformer network to conduct hand gesture recognition using high-density sEMG (HD-sEMG) signals. Taking advantage of the attention mechanism, which is incorporated into the transformer architectures, our proposed CT-HGR framework overcomes major constraints associated with most of the existing deep learning models such as model complexity;requiring feature engineering;inability to consider both temporal and spatial information of HD-sEMG signals, and;requiring a large number of training samples. The attention mechanism in the proposed model identifies similarities among different data segments with a greater capacity for parallel computations and addresses the memory limitation problems while dealing with inputs of large sequence lengths. CT-HGR can be trained from scratch without any need for transfer learning and can simultaneously extract both temporal and spatial features of HD-sEMG data. Additionally, the CT-HGR framework can perform instantaneous recognition using sEMG image spatially composed from HD-sEMG signals. A variant of the CT-HGR is also designed to incorporate microscopic neural drive information in the form of Motor Unit Spike Trains (MUSTs) extracted from HD-sEMG signals using Blind Source Separation (BSS). This variant is combined with its baseline version via a hybrid architecture to evaluate potentials of fusing macroscopic and microscopic neural drive information. The utilized HD-sEMG dataset involves 128 electrodes that collect the signals related to 65 isometric hand gestures of 20 subjects. The proposed CT-HGR framework is applied to 31.25, 62.5, 125, 250 ms window sizes of the above-mentioned dataset utilizing 32, 64, 128 electrode channels. The av

关键词： Electrodes

Dynamic Pricing based Near-Optimal Resource Allocation for Elastic Edge Offloading

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Xia, Yun Xue, Hai Zhang, Di Mumtaz, Shahid Xu, Xiaolong Rodrigues, Joel J.P.C. School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China School of Electrical and Information Engineering Zhengzhou University the Henan International Joint Laboratory of Intelligent Health Information System The National Telemedicine Center The National Engineering Laboratory for Internet Medical Systems and Applications Zhengzhou450001 China Department of Engineering Nottingham Trent University NottinghamNG1 4FQ United Kingdom The Department of Electronic Engineering Kyung Hee University Gyeonggi-do Yongin-si17104 Korea Republic of School of Software Nanjing University of Information Science and Technology Nanjing210044 China Amazonas State University AM Manaus Brazil

In mobile edge computing (MEC), task offloading can significantly reduce task execution latency and energy consumption of end user (EU). However, edge server (ES) resources are limited, necessitating efficient allocation to ensure the sustainable and healthy development for MEC systems. In this paper, we propose a dynamic pricing mechanism based near-optimal resource allocation for elastic edge offloading. First, we construct a resource pricing model and accordingly develop the utility functions for both EU and ES, the optimal pricing model parameters are derived by optimizing the utility functions. In the meantime, our theoretical analysis reveals that the EU’s utility function reaches a local maximum within the search range, but exhibits barely growth with increased resource allocation beyond this point. To this end, we further propose the Dynamic Inertia and Speed-Constrained particle swarm optimization (DISC-PSO) algorithm, which efficiently identifies the near-optimal resource allocation. Comprehensive simulation results validate the effectiveness of DISC-PSO, demonstrating that it significantly outperforms existing schemes by reducing the average number of iterations to reach a near-optimal solution by 92.11%, increasing the final user utility function value by 0.24%, and decreasing the variance of results by 95.45%. Copyright © 2024, The Authors. All rights reserved.

关键词： Particle swarm optimization (PSO)

Comparative Analysis of Machine Learning Models for Students’ Performance Prediction

学校读者我要写书评

暂无评论

Comparative Analysis of Machine Learning Models for Students...

International Conference on Advances in Digital Science, ICADS 2021

作者： Ismail, Leila Materwala, Huned Hennebelle, Alain Distributed Computing and Systems Research Laboratory Department of Computer Science and Software Engineering College of Information Technology United Arab Emirates University Al Ain Abu Dhabi15551 United Arab Emirates Al Ain Abu Dhabi United Arab Emirates

ISBN: (纸本)9783030717810

Machine learning for education is an emerging discipline where a model is developed based on training data to make predictions on students’ performance. The main aim is to identify students who would have difficulty in their learning and to take precautionary measures to help them. In this paper, we conduct a comparative analysis of the most used machine learning classification models in the literature. We evaluate the performance of the models in terms of accuracy, F-measure, and execution time using two real-life education datasets. The performance of the models is data-driven. We give insights into the models’ performance and advise on the best model to use accordingly. We believe the results of this paper will be widely used by education professionals for accurate predictions. © 2021, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Machine learning