检索结果-内蒙古大学图书馆

NTIRE 2022 Challenge on Efficient Super-Resolution: Methods and Results

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Li, Yawei Zhang, Kai Timofte, Radu Van Gool, Luc Kong, Fangyuan Li, Mingxi Liu, Songwei Du, Zongcai Liu, Ding Zhou, Chenhui Chen, Jingyi Han, Qingrui Li, Zheyuan Liu, Yingqi Chen, Xiangyu Cai, Haoming Qiao, Yu Dong, Chao Sun, Long Pan, Jinshan Zhu, Yi Zong, Zhikai Liu, Xiaoxiao Hui, Zheng Yang, Tao Ren, Peiran Xie, Xuansong Hua, Xian-Sheng Wang, Yanbo Ji, Xiaozhong Lin, Chuming Luo, Donghao Tai, Ying Wang, Chengjie Zhang, Zhizhong Xie, Yuan Cheng, Shen Luo, Ziwei Yu, Lei Wen, Zhihong Wu, Qi Li, Youwei Fan, Haoqiang Sun, Jian Liu, Shuaicheng Huang, Yuanfei Jin, Meiguang Huang, Hua Liu, Jing Zhang, Xinjian Wang, Yan Long, Lingshun Li, Gen Zhang, Yuanfan Cao, Zuowei Sun, Lei Alexander, Panaetov Wang, Yucong Cai, Minjie Wang, Li Tian, Lu Wang, Zheyuan Ma, Hongbing Liu, Jie Chen, Chao Cai, Yidong Tang, Jie Wu, Gangshan Wang, Weiran Huang, Shirui Lu, Honglei Liu, Huan Wang, Keyan Chen, Jun Chen, Shi Miao, Yuchun Huang, Zimo Zhang, Lefei Ayazoglu, Mustafa Xiong, Wei Xiong, Chengyi Wang, Fei Li, Hao Wen, Ruimian Yang, Zhijing Zou, Wenbin Zheng, Weixin Ye, Tian Zhang, Yuncheng Kong, Xiangzhen Arora, Aditya Zamir, Syed Waqas Khan, Salman Hayat, Munawar Khan, Fahad Shahbaz Gao, Dandan Zhou, Dengwen Ning, Qian Tang, Jingzhu Huang, Han Wang, Yufei Peng, Zhangheng Li, Haobo Guan, Wenxue Gong, Shenghua Li, Xin Liu, Jun Wang, Wanjun Zeng, Kun Lin, Hanjiang Chen, Xinyu Fang, Jinsheng Zhang, Shuhao Zhang, Yuhao Sinha, Abhishek Kumar Moorthi, S. Manthira Dhar, Debajyoti Yang, Hao-Hsiang Huang, Zhi-Kai Chen, Wei-Ting Chang, Hua-En Kuo, Sy-Yen Tan, Wei Chen, Hao Xu, Qian Narang, Pratik Singh, Usneek Sameen, Syed Khaitan, Harsh Yinghua, Liu Tianlin, Zhang Xiaoming, Zhang Meng, Dingxuan Tian, Chunwei Morshed, Mashrur M. Ahsan, Ahmad Omar Computer Vision Lab ETH Zurich Switzerland University of Würzburg Germany ByteDance Shenzhen China State Key Laboratory for Novel Software Technology Nanjing University China ByteDance Inc China NetEase Inc. China Shenzhen Institutes of Advanced Technology CAS China University of Macau China Shanghai AI Lab Shanghai China Nanjing University of Science and Technology China Amazon Web Services United States China Alibaba DAMO Academy EFC Yuhang District Zhejiang Hangzhou China East China Normal University China Youtu Lab Tencent China Megvii Technology China University of Electronic Science and Technology of China China School of Artificial Intelligence Beijing Normal University China Alibaba Group China Bilibili AI China Nankai-Baidu Joint Lab Nankai University Tianjin China Platform Technologies Tencent Online Video China Higher School of Economics Russia Huawei Moscow Research Center Russia Hunan University China Xidian University China Xilinx Technology Beijing Limited China College of Information Science and Engineering Xinjiang University Urumqi China Department of Electronic Engineering Tsinghua University Beijing China Nanjing University China McMaster University Canada School of Telecommunication Engineering Xidian University Xi’an China School of Computer Science Wuhan University Wuhan China School of Mathematical Science University of Electronic Science and Technology of China Chengdu China School of Computer Science The University of Sydney Sydney Australia Aselsan Research Ankara Turkey School of Electronic and Information Engineering South-Central University for Nationalities Wuhan China Guangdong University of Technology China Fujian Normal University Fuzhou University Jimei University China Design Group China Abu Dhabi United Arab Emirates Monash University Melbourne Australia Mohamed bin Zayed University of AI United Arab Emirates North China Electric Power University Changping District Beijing China The School of Art

This paper reviews the NTIRE 2022 challenge on efficient single image super-resolution with focus on the proposed solutions and results. The task of the challenge was to super-resolve an input image with a magnification factor of ×4 based on pairs of low and corresponding high resolution images. The aim was to design a network for single image super-resolution that achieved improvement of efficiency measured according to several metrics including runtime, parameters, FLOPs, activations, and memory consumption while at least maintaining the PSNR of 29.00dB on DIV2K validation set. IMDN is set as the baseline for efficiency measurement. The challenge had 3 tracks including the main track (runtime), sub-track one (model complexity), and sub-track two (overall performance). In the main track, the practical runtime performance of the submissions was evaluated. The rank of the teams were determined directly by the absolute value of the average runtime on the validation set and test set. In sub-track one, the number of parameters and FLOPs were considered. And the individual rankings of the two metrics were summed up to determine a final ranking in this track. In sub-track two, all of the five metrics mentioned in the description of the challenge including runtime, parameter count, FLOPs, activations, and memory consumption were considered. Similar to sub-track one, the rankings of five metrics were summed up to determine a final ranking. The challenge had 303 registered participants, and 43 teams made valid submissions. They gauge the state-of-the-art in efficient single image super-resolution. © 2022, CC BY.

关键词： Chemical activation

Unified quality assessment of in-the-wild videos with mixed datasets training

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Li, Dingquan Jiang, Tingting Jiang, Ming National Engineering Laboratory for Video Technology Peking University Beijing China Advanced Institute of Information Technology Peking University Hangzhou China Department of Computer Science Peking University Beijing China Laboratory of Mathematics and Its Applications School of Mathematical Sciences Peking University Beijing China Beijing International Center for Mathematical Research Peking University Beijing China Advanced Innovation Center for Future Visual Entertainment Beijing Film Academy Beijing China

Video quality assessment (VQA) is an important problem in computer vision. The videos in computer vision applications are usually captured in the wild. We focus on automatically assessing the quality of in-the-wild videos, which is a challenging problem due to the absence of reference videos, the complexity of distortions, and the diversity of video contents. Moreover, the video contents and distortions among existing datasets are quite different, which leads to poor performance of data-driven methods in the cross-dataset evaluation setting. To improve the performance of quality assessment models, we borrow intuitions from human perception, specifically, content dependency and temporal-memory effects of human visual system. To face the cross-dataset evaluation challenge, we explore a mixed datasets training strategy for training a single VQA model with multiple datasets. The proposed unified framework explicitly includes three stages: relative quality assessor, nonlinear mapping, and dataset-specific perceptual scale alignment, to jointly predict relative quality, perceptual quality, and subjective quality. Experiments are conducted on four publicly available datasets for VQA in the wild, i.e., LIVE-VQC, LIVE-Qualcomm, KoNViD-1k, and CVD2014. The experimental results verify the effectiveness of the mixed datasets training strategy and prove the superior performance of the unified model in comparison with the state-of-the-art models. For reproducible research, we make the PyTorch implementation of our method available at https://***/lidq92/MDTVSFA. Copyright © 2020, The Authors. All rights reserved.

关键词： Video recording

A Distributed Privacy-Preserving Learning Dynamics in General Social Networks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Tao, Youming Chen, Shuzhen Li, Feng Yu, Dongxiao Yu, Jiguo Sheng, Hao School of Computer Science and Technology Shandong University Qingdao266237 China Big Data Institute Qilu University of Technology Jinan250353 China State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang University Beijing100191 China Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing100191 China

In this paper, we study a distributed privacy-preserving learning problem in social networks with general topology. The agents can communicate with each other over the network, which may result in privacy disclosure, since the trustworthiness of the agents cannot be guaranteed. Given a set of options which yield unknown stochastic rewards, each agent is required to learn the best one, aiming at maximizing the resulting expected average cumulative reward. To serve the above goal, we propose a four-staged distributed algorithm which efficiently exploits the collaboration among the agents while preserving the local privacy for each of them. In particular, our algorithm proceeds iteratively, and in every round, each agent i) randomly perturbs its adoption for the privacy-preserving purpose, ii) disseminates the perturbed adoption over the social network in a nearly uniform manner through random walking, iii) selects an option by referring to the perturbed suggestions received from its peers, and iv) decides whether or not to adopt the selected option as preference according to its latest reward feedback. Through solid theoretical analysis, we quantify the trade-off among the number of agents (or communication overhead), privacy preserving and learning utility. We also perform extensive simulations to verify the efficacy of our proposed social learning algorithm. Copyright © 2020, The Authors. All rights reserved.

关键词： Iterative methods

DPA-2: a large atomic model as a multi-task learner

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Duo Liu, Xinzijian Zhang, Xiangyu Zhang, Chengqian Cai, Chun Bi, Hangrui Du, Yiming Qin, Xuejian Peng, Anyang Huang, Jiameng Li, Bowen Shan, Yifan Zeng, Jinzhe Zhang, Yuzhi Liu, Siyuan Li, Yifan Chang, Junhan Wang, Xinyan Zhou, Shuo Liu, Jianchuan Luo, Xiaoshan Wang, Zhenyu Jiang, Wanrun Wu, Jing Yang, Yudi Yang, Jiyuan Yang, Manyi Gong, Fu-Qiang Zhang, Linshuang Shi, Mengchao Dai, Fu-Zhi York, Darrin M. Liu, Shi Zhu, Tong Zhong, Zhicheng Lv, Jian Cheng, Jun Jia, Weile Chen, Mohan Ke, Guolin Weinan, E. Zhang, Linfeng Wang, Han AI for Science Institute Beijing100080 China DP Technology Beijing100080 China Academy for Advanced Interdisciplinary Studies Peking University Beijing100871 China State Key Lab of Processors Institute of Computing Technology Chinese Academy of Sciences Beijing100871 China University of Chinese Academy of Sciences Beijing100871 China HEDPS CAPT College of Engineering Peking University Beijing100871 China Ningbo Institute of Materials Technology and Engineering Chinese Academy of Sciences Ningbo315201 China CAS Key Laboratory of Magnetic Materials and Devices Zhejiang Province Key Laboratory of Magnetic Materials and Application Technology Chinese Academy of Sciences Ningbo315201 China School of Electronics Engineering and Computer Science Peking University Beijing100871 China Shanghai Engineering Research Center of Molecular Therapeutics & New Drug Development School of Chemistry and Molecular Engineering East China Normal University Shanghai200062 China Laboratory for Biomolecular Simulation Research Institute for Quantitative Biomedicine Department of Chemistry and Chemical Biology Rutgers University PiscatawayNJ08854 United States Department of Chemistry Princeton University PrincetonNJ08540 United States College of Chemistry and Molecular Engineering Peking University Beijing100871 China Yuanpei College Peking University Beijing100871 China School of Electrical Engineering and Electronic Information Xihua University Chengdu610039 China State Key Laboratory of Superhard Materials College of Physics Jilin University Changchun130012 China Key Laboratory of Material Simulation Methods & Software of Ministry of Education College of Physics Jilin University Changchun130012 China International Center of Future Science Jilin University Changchun130012 China Key Laboratory for Quantum Materials of Zhejiang Province Department of Physics School of Science Westlake University Zhejiang Hangzhou310030 China Atomistic Simulations Italia

The rapid advancements in artificial intelligence (AI) are catalyzing transformative changes in atomic modeling, simulation, and design. AI-driven potential energy models have demonstrated the capability to conduct large-scale, long-duration simulations with the accuracy of ab initio electronic structure methods. However, the model generation process remains a bottleneck for large-scale applications. We propose a shift towards a model-centric ecosystem, wherein a large atomic model (LAM), pre-trained across multiple disciplines, can be efficiently fine-tuned and distilled for various downstream tasks, thereby establishing a new framework for molecular modeling. In this study, we introduce the DPA-2 architecture as a prototype for LAMs. Pre-trained on a diverse array of chemical and materials systems using a multi-task approach, DPA-2 demonstrates superior generalization capabilities across multiple downstream tasks compared to the traditional single-task pre-training and fine-tuning methodologies. Our approach sets the stage for the development and broad application of LAMs in molecular and materials simulation research. Copyright © 2023, The Authors. All rights reserved.

关键词： Multi-task learning

Rely-Guarantee Reasoning about Messaging System for Autonomous Vehicles

学校读者我要写书评

暂无评论

Rely-Guarantee Reasoning about Messaging System for Autonomo...

Theoretical Aspects of software engineering (TASE)

作者： Wenjing Xu Yongwang Zhao Dianfu Ma YuXin Zhang Qian Xiao State Key Laboratory of Software Development Environment School of Computer Science and Engineering Beihang Univerisity Beijing China Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing China State Key Laboratory of Automotive Simulation and Control College of Automotive Engineering Jilin University Jilin China School of Computer Science and Technology Guizhou University Guiyang China

Messaging system as a communicating infrastructure is a safety-critical component of autonomous vehicles. For the purpose of safety certification of Level 4 autonomous driving systems, its necessary to provide a formally verified specification of messaging systems. This paper presents a realistic case study using the PiCore rely-guarantee framework to formally verify the DGPS (Differential Global Positioning System) of UISEE autonomous driving systems. We first create an axiom model of messaging systems by extending PiCore, which is reusable for concrete applications. The model supports dynamic configuration of message buffers as well as the automatic generation of the rely and guarantee conditions for compositional reasoning. It is instantiated when developing the formal specification of DGPS. Then, we use the rely-guarantee proof system of PiCore to verify the functional correctness and invariant of DGPS. The specification and its proof provide a strong evidence for the ongoing safety certification of DGPS.

关键词： Cognition Safety Formal specifications Vehicle dynamics Autonomous vehicles Certification Global Positioning System

Energy Efficient User Clustering, Hybrid Precoding and Power Optimization in Terahertz MIMO-NOMA Systems

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhang, Haijun Zhang, Haisen Liu, Wei Long, Keping Dong, Jiangbo Leung, Victor C.M. Institute of Artificial Intelligence Beijing Advanced Innovation Center for Materials Genome Engineering Beijing Engineering and Technology Research Center for Convergence Networks and Ubiquitous Services University of Science and Technology Beijing Beijing100083 China China Mobile Group Design Institute Co. Ltd Beijing100080 China College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Electrical and Computer Engineering University of British Columbia VancouverBC V6T 1Z4 Canada

Terahertz (THz) band communication has been widely studied to meet the future demand for ultra-high capacity. In addition, multi-input multi-output (MIMO) technique and non-orthogonal multiple access (NOMA) technique with multi-antenna also enable the network to carry more users and provide multiplexing gain. In this paper, we study the maximization of energy efficiency (EE) problem in THz-NOMA-MIMO systems for the first time. And the original optimization problem is divided into user clustering, hybrid precoding and power optimization. Based on channel correlation characteristics, a fast convergence scheme for user clustering in THz-NOMA-MIMO system using enhanced K-means machine learning algorithm is proposed. Considering the power consumption and implementation complexity, the hybrid precoding scheme based on the sub-connection structure is adopted. Considering the fronthaul link capacity constraint, we design a distributed alternating direction method of multipliers (ADMM) algorithm for power allocation to maximize the EE of THz-NOMA cache-enabled system with imperfect successive interference cancellation (SIC). The simulation results show that the proposed user clustering scheme can achieve faster convergence and higher EE, the design of the hybrid precoding of the sub-connection structure can achieve lower power consumption and power optimization can achieve a higher EE for the THz cache-enabled network. Copyright © 2020, The Authors. All rights reserved.

关键词： MIMO systems

Unsupervised feature selection via adaptive embedding and sparse learning for Parkinson’s disease diagnosis 1

学校读者我要写书评

暂无评论

3rd International Workshop on Connectomics in NeuroImaging, CNI 2019 held in conjunction with the 22nd International Conference on Medical Image Computing and computer Assisted Intervention, MICCAI 2019

作者： Huang, Zhongwei Lei, Haijun Chen, Guoliang Li, Shiqi Li, Hancong Elazab, Ahmed Lei, Baiying Key Laboratory of Service Computing and Applications Guangdong Province Key Laboratory of Popular High Performance Computers College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China National-Regional Key Technology Engineering Laboratory for Medical Ultrasound Guangdong Key Laboratory for Biomedical Measurements and Ultrasound Imaging School of Biomedical Engineering Health Science Center Shenzhen University Shenzhen518060 China

ISBN: (数字)9783030323912

ISBN: (纸本)9783030323905

Parkinson’s disease (PD) is known as a progressive neurodegenerative disease in elderly people. Apart from decelerating the disease exacerbation, early and accurate diagnosis also alleviates mental and physical sufferings and provides timely and appropriate medication. In this paper, we propose an unsupervised feature selection method via adaptive manifold embedding and sparse learning exploiting longitudinal multimodal neuroimaging data for classification and regression prediction. Specifically, the proposed method simultaneously carries out feature selection and dynamic local structure learning to obtain the structural information inherent in the neuroimaging data. We conduct extensive experiments on the publicly available Parkinson’s progression markers initiative (PPMI) dataset to validate the proposed method. Our proposed method outperforms other state-of-the-art methods in terms of classification and regression prediction performance. © Springer Nature Switzerland AG 2019.

关键词： Neuroimaging

Adversarial directed graph embedding

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhu, Shijie Li, Jianxin Peng, Hao Wang, Senzhang He, Lifang Beijing Advanced Innovation Center for Big Data and Brain Computing Beihang University Beijing100191 China State Key Laboratory of Software Development Environment Beihang University Beijing100191 China College of Computer Science and Technology Nanjing University of Aeronautics and Astronautics Nanjing211106 China Department of Computer Science and Engineering Lehigh University BethlehemPA United States

Node representation learning for directed graphs is critically important to facilitate many graph mining tasks. To capture the directed edges between nodes, existing methods mostly learn two embedding vectors for each node, source vector and target vector. However, these methods learn the source and target vectors separately. For the node with very low indegree or outdegree, the corresponding target vector or source vector cannot be effectively learned. In this paper, we propose a novel Directed Graph embedding framework based on Generative Adversarial Network, called DGGAN. The main idea is to use adversarial mechanisms to deploy a discriminator and two generators that jointly learn each node's source and target vectors. For a given node, the two generators are trained to generate its fake target and source neighbor nodes from the same underlying distribution, and the discriminator aims to distinguish whether a neighbor node is real or fake. The two generators are formulated into a unified framework and could mutually reinforce each other to learn more robust source and target vectors. Extensive experiments show that DGGAN consistently and significantly outperforms existing state-of-the-art methods across multiple graph mining tasks on directed graphs. Copyright © 2020, The Authors. All rights reserved.

关键词： Directed graphs

A Review of Artificial Fish Swarm Algorithms: Recent Advances and applications

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Pourpanah, Farhad Wang, Ran Lim, Chee Peng Wang, Xi-Zhao Yazdani, Danial College of Mathematics and Statistics Guangdong Key Lab. of Intelligent Information Processing Shenzhen University China Department of Electrical and Computer Engineering University of Windsor Canada College of Mathematics and Statistics Shenzhen Key Lab. of Advanced Machine Learning and Applications Guangdong Key Lab. of Intelligent Information Processing Shenzhen University China Institute for Intelligent Systems Research and Innovation Deakin University Australia College of Computer Science and Software Engineering Guangdong Key Lab. of Intelligent Information Processing Shenzhen University China School of Computer Science and Engineering Southern University of Science and Technology China

The Artificial Fish Swarm Algorithm (AFSA) is inspired by the ecological behaviors of fish schooling in nature, viz., the preying, swarming and following behaviors. Owing to a number of salient properties, which include flexibility, fast convergence, and insensitivity to the initial parameter settings, the family of AFSA has emerged as an effective Swarm Intelligence (SI) methodology that has been widely applied to solve real-world optimization problems. Since its introduction in 2002, many improved and hybrid AFSA models have been developed to tackle continuous, binary, and combinatorial optimization problems. This paper aims to present a concise review of the continuous AFSA, encompassing the original ASFA, its improvements and hybrid models, as well as their associated applications. We focus on articles published in high-quality journals since 2013. Our review provides insights into AFSA parameters modifications, procedure and sub-functions. The main reasons for these enhancements and the comparison results with other hybrid methods are discussed. In addition, hybrid, multi-objective and dynamic AFSA models that have been proposed to solve continuous optimization problems are elucidated. We also analyse possible AFSA enhancements and highlight future research directions for advancing AFSA-based models. Copyright © 2020, The Authors. All rights reserved.

关键词： Multiobjective optimization