检索结果-南通市图书馆

Learning practically feasible policies for online 3D bin packing

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2022年第1期65卷 175-191页

作者： Hang ZHAO Chenyang ZHU Xin XU Hui HUANG Kai XU School of Computer Science National University of Defense Technology College of Intelligence Science and Technology National University of Defense Technology College of Computer Science and Software Engineering Shenzhen University

We tackle the online 3 D bin packing problem(3 D-BPP), a challenging yet practically useful variant of the classical bin packing problem. In this problem, the items are delivered to the agent without informing the full sequence information. The agent must directly pack these items into the target bin stably without changing their arrival order, and no further adjustment is permitted. Online 3 D-BPP can be naturally formulated as a Markov decision process(MDP). We adopt deep reinforcement learning, in particular, the on-policy actor-critic framework, to solve this MDP with constrained action space. To learn a practically feasible packing policy, we propose three critical designs. First, we propose an online analysis of packing stability based on a novel stacking tree. It attains a high analysis accuracy while reducing the computational complexity from O(N2) to O(N log N), making it especially suited for reinforcement learning training. Second, we propose a decoupled packing policy learning for different dimensions of placement which enables high-resolution spatial discretization and hence high packing precision. Third, we introduce a reward function that dictates the robot to place items in a far-to-near order and therefore simplifies the collision avoidance in movement planning of the robotic arm. Furthermore, we provide a comprehensive discussion on several key implemental issues. The extensive evaluation demonstrates that our learned policy outperforms the state-of-the-art methods significantly and is practically usable for real-world applications.

关键词： bin packing problem online 3D-BPP reinforcement learning

Efficient deep blue OLEDs with extremely low efficiency roll-off at high brightness based on phenanthroimidazole derivatives

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Chinese Chemical Letters 2019年第11期30卷 1989-1993页

作者： Xiang Chen Jue-Wen Zhao Xu-Hui Zheng Jie-Ji Zhu Guo-Xi Yang Shan-Shun Tang Qing-Xiao Tong Si-Lu Tao Department of Chemistry and key Laboratory for Preparation and Application of Ordered Structural Material of Guangdong Province Shantou UniversityGuangdong 515063China School of Optoelectronic Science and Engineering University of Electronic Science and Technology of China(UESTC)Chengdu 610054China

Phosphorescent and thermally activated delayed fluorescence(TADF)emitters can break through the spin statistics rules and achieve great success in external quantum efficiency(over 5%).However,maintaining high efficiency at high brightness is a tremendous challenge for applications of organic light emitting ***,we reported two phenanthroimidazole derivatives PPI-An-CN and PPI-An-TP and achieved extremely low efficiency roll-off with about 99%of the maximum external quantum efficiency(EQEmax)maintained even at a high luminance of 1000 cd/cm2 based non-doped *** doping the two materials in CBP(4,4'-bis(N-carbazolyl)-1,1'-biphenyl),the doped devices still exhibited excellent stability at high brightness with CIEy≈0.07 and low turn-on voltage of only 2.8 *** state-ofthe-art low efficiency roll-off makes the new materials attractive for potential *** is the first time that the Fragment Contribution Analysis method has been used to analyze the excited state properties of the molecules in the field of OLEDs,which helps us understand the mechanism more intuitively and deeply.

关键词： OLED Deep blue Phenanthroimidazole Low efficiency roll-off Fragment contribution analysis

维普期刊数据库

Journal of Hepatology|代谢生物标志物显著增强了对HBV相关慢加急性肝衰竭发生和结果的预测

在线全文

学校读者我要写书评

暂无评论

临床肝胆病杂志 2024年第2期40卷 257-257页

作者：冯悦高沿航吉林大学第一医院感染病与病原生物学中心肝胆胰内科

慢加急性肝衰竭(ACLF)是一种与慢性肝病患者高短期病死率相关的临床综合征。慢性乙型肝炎是中国和亚洲其他国家ACLF(HBV-ACLF)的主要病因。为了提高ACLF患者的疾病管理和生存率,来自上海交通大学的张艳等进行了一项多中心前瞻性研究,旨... 详细信息

慢加急性肝衰竭(ACLF)是一种与慢性肝病患者高短期病死率相关的临床综合征。慢性乙型肝炎是中国和亚洲其他国家ACLF(HBV-ACLF)的主要病因。为了提高ACLF患者的疾病管理和生存率,来自上海交通大学的张艳等进行了一项多中心前瞻性研究,旨在发现新的生物标志物,以增强对HBV-ACLF的诊断和预测。该研究对1 024份血浆样本进行了代谢组学分析,这些样本来自入院时出现急性加重的HBV相关慢性肝病患者(367份ACLF和657份非ACLF)。将样本随机分为相等的两份,作为发现集和验证集。

关键词：慢性肝病患者临床综合征上海交通大学生物标志物血浆样本慢性乙型肝炎急性加重主要病因

维普期刊数据库博看期刊

Inferring object properties from human interaction and transferring them to new motions

在线全文

学校读者我要写书评

暂无评论

Computational Visual Media 2021年第3期7卷 375-392页

作者： Qian Zheng Weikai Wu Hanting Pan Niloy Mitra Daniel Cohen-Or Hui Huang Shenzhen University ShenzhenChina University College London UK. Tel Aviv University Tel-AvivIsrael

Humans regularly interact with their surrounding *** interactions often result in strongly correlated motions between humans and the interacting *** thus ask:"Is it possible to infer object properties from skeletal motion alone,even without seeing the interacting object itself?"In this paper,we present a fine-grained action recognition method that learns to infer such latent object properties from human interaction motion *** inference allows us to disentangle the motion from the object property and transfer object properties to a given *** collected a large number of videos and 3 D skeletal motions of performing actors using an inertial motion capture *** analyzed similar actions and learned subtle differences between them to reveal latent properties of the interacting *** particular,we learned to identify the interacting object,by estimating its weight,or its *** results clearly demonstrate that motions and interacting objects are highly correlated and that related object latent properties can be inferred from 3 D skeleton sequences alone,leading to new synthesis possibilities for motions involving human *** dataset is available at http://***/research/2020/***.

关键词： human interaction motion object property inference motion analysis motion synthesis

维普期刊数据库

On learning the right attention point for feature enhancement

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2023年第1期66卷 131-143页

作者： Liqiang LIN Pengdi HUANG Chi-Wing FU Kai XU Hao ZHANG Hui HUANG College of Computer Science and Software Engineering Shenzhen University Department of Computer Science and Engineering The Chinese University of Hong Kong School of Computer Science National University of Defense Technology School of Computing Science Simon Fraser University

We present a novel attention-based mechanism to learn enhanced point features for point cloud processing tasks, e.g., classification and segmentation. Unlike prior studies, which were trained to optimize the weights of a pre-selected set of attention points, our approach learns to locate the best attention points to maximize the performance of a specific task, e.g., point cloud classification. Importantly, we advocate the use of single attention point to facilitate semantic understanding in point feature learning. Specifically,we formulate a new and simple convolution, which combines convolutional features from an input point and its corresponding learned attention point(LAP). Our attention mechanism can be easily incorporated into state-of-the-art point cloud classification and segmentation networks. Extensive experiments on common benchmarks, such as Model Net40, Shape Net Part, and S3DIS, all demonstrate that our LAP-enabled networks consistently outperform the respective original networks, as well as other competitive alternatives, which employ multiple attention points, either pre-selected or learned under our LAP framework.

关键词： point convolution feature enhancement attention point deep neural network

Virtual sample generation for model-based prognostics and health management of on-board high-speed train control system

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

High-Speed Railway 2023年第3期1卷 153-161页

作者： Jiang Liu Baigen Cair Jinlan Wang Jian Wang School of Electronic and Information Engineering Beijing Jiaotong UniversityBeijing 100044China Frontiers Science Center for Smart High-speed Railway System Bejing Jiaotong UniversityBeijing 100044China School of Computer and Information Technology Bejing Jiaotong UniversityBejing 100044China Guangzhou Railway Polytechnic Guangzhou 511300China

In view of class imbalance in data-driven modeling for Prognostics and Health Management(PHM),existing classification methods may fail in generating effective fault prediction models for the on-board high-speed train control equipment.A virtual sample generation solution based on Generative Adversarial Network(GAN)is proposed to overcome this *** at augmenting the sample classes with the imbalanced data problem,the GAN-based virtual sample generation strategy is embedded into the establishment of fault prediction *** the PHM framework of the on-board train control system,the virtual sample generation principle and the detailed procedures are *** the enhanced class-balancing mechanism and the designed sample augmentation logic,the PHM scheme of the on-board train control equipment has powerful data condition adaptability and can effectively predict the fault probability and life cycle *** data from a specific type of on-board train control system is employed for the validation of the presented *** comparative results indicate that GAN-based sample augmentation is capable of achieving a desirable sample balancing level and enhancing the performance of correspondingly derived fault prediction models for the Condition-based Maintenance(CBM)operations.

关键词： High-speed railway Prognostics and health management Train control Virtual sample Generative adversarial network

维普期刊数据库

Reflecting Surface-Aided Device-to-Device Communications System

在线全文

学校读者我要写书评

暂无评论

Journal of Communications and Information Networks 2020年第4期5卷 403-410页

作者： Chiya Zhang Wenyu Chen Chunlong He Xingquan Li School of Electronic and Information Engineering Harbin Institute of TechnologyShenzhen 518055China.Peng Cheng Laboratory(PCL)Shenzhen 5180523China Guangdong key Laboratory of Intelligent Information Processing Shenzhen UniversityShenzhen518060China

Intelligent reflecting surface(IRS)is a promising technology for its capability of reflecting the incident signal towards the desired *** can improve the efficiency of wireless communication *** paper introduces IRS into a device-to-device(D2D)communications system to improve the throughput of the D2D *** adopt the block coordinate descent al-gorithm and semidefinite relaxation technique to optimize the beamforming vector,power allocation and phase shift *** results demonstrate that IRS is able to enhance the throughput of the D2D communications system,and the proposed algorithm significantly outper-forms the other benchmark algorithms.

关键词： intelligent reflecting surface device-to-device communications block coordinate descent

Distributed Intelligent Reflecting Surfaces-Aided Device-to-Device Communications System

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Communications and Information Networks 2021年第3期6卷 197-207页

作者： Chiya Zhang Wenyu Chen Qian Chen Chunlong He School of Electronic and Information Engineering Harbin Institute of TechnologyShenzhen 518055China Peng Cheng Laboratory(PCL) Shenzhen 5180523China Guangdong key Laboratory of Intelligent Information Processing Shenzhen UniversityShenzhen 518060China National Mobile Communications Research Laboratory Southeast UniversityNanjing 211189China

Intelligent reflecting surface(IRS)is a revolutionizing and promising technology to improve the high transmission rate of the wireless communication ***,an IRS consists of a great amount of low-cost passive reflecting elements,which reflect the incident signals to the desired user by collaboratively using passive *** paper introduces IRSs into a device-to-device(D2D)underlying cellular system to enhance transmission rate performance of the D2D *** formulate an optimization problem of maximizing the transmission rate of the D2D pairs while satisfying the minimum required rate of the cellular *** address this problem by jointly optimizing the reuse indicator,received beamforming,power allocation,and phase shift *** coordinate descent(BCD)algorithm is adopted to decouple the original problem into four *** form solutions are obtained by solving the sub-problems of optimizing the received beamforming and power ***,Kuhn-Munkres(KM)algorithm and minimization-majorization(MM)algorithm are adopted to solve the sub-problems of optimizing the reuse indicator and phase shift matrices,*** results demonstrate that IRSs can effectively improve the transmission rate of the D2D pairs and our proposed distributed IRSs scheme outperforms the other benchmark schemes.

关键词： intelligent reflecting surface device-to-device communications block coordinate descent

Hausdorff point convolution with geometric priors

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2021年第11期64卷 67-79页

作者： Liqiang LIN Pengdi HUANG Fuyou XUE Kai XU Daniel COHEN-OR Hui HUANG College of Computer Science and Software Engineering Shenzhen University School of Computer Science National University of Defense Technology

Developing point convolution for irregular point clouds to extract deep features remains challenging. Current methods evaluate the response by computing point set distances which account only for the spatial alignment between two point sets, but not quite for their underlying shapes. Without a shapeaware response, it is hard to characterize the 3 D geometry of a point cloud efficiently with a compact set of kernels. In this paper, we advocate the use of modified Hausdorff distance as a shape-aware distance measure for calculating point convolutional responses. The technique we present, coined Hausdorff point convolution(HPC), is shape-aware. We show that HPC constitutes a powerful point feature learning with a rather compact set of only four types of geometric priors as kernels. We further develop an HPC-based deep neural network(HPC-DNN). Task-specific learning can be achieved by tuning the network weights for combining the shortest distances between the input and the kernel point sets. We also realize hierarchical feature learning by designing a multi-kernel HPC for multi-scale feature encoding. Extensive experiments demonstrate that HPC-DNN outperforms strong point convolution baselines(e.g., KPConv), achieving 2.8%m Io U performance boost on S3 DIS and 1.5% on Semantic KITTI for semantic segmentation task.

关键词： point convolution Hausdorff distance geometric prior deep neural network