检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Gao, Jiaqi Huang, Zhizhong Lei, Yiming Shan, Hongming Wang, James Z. Wang, Fei-Yue Zhang, Junping The Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China The Institute of Science and Technology for Brain-inspired Intelligence MOE Frontiers Center for Brain Science Fudan University Shanghai200433 China The Shanghai Center for Brain Science and Brain-Inspired Technology Shanghai201210 China The College of Information Sciences and Technology The Pennsylvania State University University ParkPA16802 United States The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The Institute of Systems Engineering Macau University of Science and Technology China Qingdao Academy of Intelligent Industries Qingdao266109 China

Most conventional crowd counting methods utilize a fully-supervised learning framework to establish a mapping between scene images and crowd density maps. They usually rely on a large quantity of costly and time-intensive pixel-level annotations for training supervision. One way to mitigate the intensive labeling effort and improve counting accuracy is to leverage large amounts of unlabeled images. This is attributed to the inherent self-structural information and rank consistency within a single image, offering additional qualitative relation supervision during training. Contrary to earlier methods that utilized the rank relations at the original image level, we explore such rank-consistency relation within the latent feature spaces. This approach enables the incorporation of numerous pyramid partial orders, strengthening the model representation capability. A notable advantage is that it can also increase the utilization ratio of unlabeled samples. Specifically, we propose a Deep Rank-consistEnt pyrAmid Model (DREAM), which makes full use of rank consistency across coarse-to-fine pyramid features in latent spaces for enhanced crowd counting with massive unlabeled images. In addition, we have collected a new unlabeled crowd counting dataset, FUDAN-UCC, comprising 4,000 images for training purposes. Extensive experiments on four benchmark datasets, namely UCF-QNRF, ShanghaiTech PartA and PartB, and UCF-CC-50, show the effectiveness of our method compared with previous semi-supervised methods. The codes are available at https://***/bridgeqiqi/DREAM. Copyright © 2022, The Authors. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Development of GPU Enhanced AI Education Platforms for K-12 Schools

Development of GPU Enhanced AI Education Platforms for K-12 ...

引用

Chinese automation Congress (CAC)

作者： Xiaoyan Gong Yifan Wen Xiwei Liu Hongxia Zhao Sifeng Jing Qiang Li Kai Gao Jilong Luo The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China Qingdao Academy of Intelligent Industries Institute of Smart Education Systems Qingdao China The 21st Century International School Beijing China Beijing NO. 2 Middle School Beijing China Beijing University of Chemical Technology Beijing China

ISBN: (纸本)9781665426480

Recently, with continuous implementation of education reform, AI (Artificial Intelligence) has been included in K-12 (Kindergarten to 12th grade) curricula and AI education is in full swing in China. However, standard, cost-efficient AI education platforms for K-12 schools are still not available. So in this paper, based on Jetson Nano, we developed a GPU-enhanced AI teaching platform including AI learning computers for students to develop AI-related algorithms and applications, AI computer clusters for whole classes, and modular auto-driving cars to run and test algorithms and applications.

关键词： Computers automation Education Clustering algorithms Graphics processing units Automobiles Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Attention based Graph Covolution Networks for Intelligent Traffic Flow Analysis

Attention based Graph Covolution Networks for Intelligent Tr...

引用

IEEE International Conference on automation Science and Engineering (CASE)

作者： Hongxin Zhang Jiaxin Liu Ying Tang Gang Xiong State Key Lab of CAD & CG Zhejiang University China College of Software Engineering Zhejiang University of Technology China State Key Laboratory of Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences

ISBN: (数字)9781728169040

ISBN: (纸本)9781728169057

Predicting traffic flow is one of the fundamental issue to smart cities. However it is still challenging in vehicular cyber-physical systems because of ever-increasing urban traffic data. Most previous models for traffic flow prediction suffers the problem with variance error depends on time and location. In this paper, a novel graph convolution network (GCN) approach is proposed for predicting citywide traffic flow. Our key idea is to introduce spatial and temporal attention to the GCN model to lessen the impact of urban data complexity. A framework of flow-based graph convolutional network is established to improve traffic flow prediction while investigating the spatial and temporal correlation of traffic flow. We further propose practical strategies that efficiently learn parameters of the model. Experimental results demonstrate that the proposed approach to traffic flow prediction outperforms state-of-the-art approaches.

关键词： Predictive models Data models Analytical models Convolution Urban areas Chebyshev approximation Roads

来源：评论

学校读者我要写书评

暂无评论

An effective deep reinforcement learning approach for adaptive traffic signal control

An effective deep reinforcement learning approach for adapti...

引用

Chinese automation Congress (CAC)

作者： Mingrui Yu Jaijun Chai Yisheng Lv Gang Xiong School of Electronic and Information Engineering Xi’an Jiaotong University Xi’an China State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China

ISBN: (数字)9781728176871

ISBN: (纸本)9781728176888

Intelligent traffic signal timing is critical to reduce traffic congestion and vehicle delay. Recent studies have shown promising results of deep reinforcement learning for traffic signal control. However, existing studies have only focused on selecting which direction (phase) to let vehicles go, not on phase duration. In this paper, we propose a deep reinforcement learning algorithm that automatically learns an optimal policy to adaptively determine phase duration. To improve algorithm performance and stability, we propose a phase sensitive neural network structure based on the deep deterministic policy gradient (DDPG) model, i.e. we design a deep neural network controller for each specific traffic signal phase with DDPG; we develop some interesting training techniques to improve training efficiency, i.e. dividing the training process into three stages and introducing the episode-break mechanism. We test the proposed methods on an isolated intersection under diverse traffic demands. Experiments show that our method is more effective.

关键词： Training Mathematical model Neural networks Reinforcement learning Vehicle dynamics Stability analysis Numerical stability

来源：评论

学校读者我要写书评

暂无评论

Attention-Guided Lightweight Network for Real-Time Segmentation of Robotic Surgical Instruments

Attention-Guided Lightweight Network for Real-Time Segmentat...

引用

IEEE International Conference on Robotics and automation (ICRA)

作者： Zhen-Liang Ni Gui-Bin Bian Zeng-Guang Hou Xiao-Hu Zhou Xiao-Liang Xie Zhen Li State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China school of Artificial Intelligence University of Chinese Academy of Sciences Beijing China

ISBN: (数字)9781728173955

ISBN: (纸本)9781728173962

The real-time segmentation of surgical instruments plays a crucial role in robot-assisted surgery. However, it is still a challenging task to implement deep learning models to do real-time segmentation for surgical instruments due to their high computational costs and slow inference speed. In this paper, we propose an attention-guided lightweight network (LWANet), which can segment surgical instruments in real-time. LWANet adopts encoder-decoder architecture, where the encoder is the lightweight network MobileNetV2, and the decoder consists of depthwise separable convolution, attention fusion block, and transposed convolution. Depthwise separable convolution is used as the basic unit to construct the decoder, which can reduce the model size and computational costs. Attention fusion block captures global contexts and encodes semantic dependencies between channels to emphasize target regions, contributing to locating the surgical instrument. Transposed convolution is performed to upsample feature maps for acquiring refined edges. LWANet can segment surgical instruments in real-time while takes little computational costs. Based on 960x544 inputs, its inference speed can reach 39 fps with only 3.39 GFLOPs. Also, it has a small model size and the number of parameters is only 2.06 M. The proposed network is evaluated on two datasets. It achieves state-of-the- art performance 94.10% mean IOU on Cata7 and obtains a new record on EndoVis 2017 with a 4.10% increase on mean IOU.

关键词： Instruments Convolution Semantics Computational efficiency Decoding Surgery Real-time systems

来源：评论

学校读者我要写书评

暂无评论

Cooperative Multi-Agent Deep Reinforcement Learning with Counterfactual Reward

Cooperative Multi-Agent Deep Reinforcement Learning with Cou...

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Kun Shao Yuanheng Zhu Zhentao Tang Dongbin Zhao State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China

ISBN: (数字)9781728169262

ISBN: (纸本)9781728169279

In partially observable fully cooperative games, agents generally tend to maximize global rewards with joint actions, so it is difficult for each agent to deduce their own contribution. To address this credit assignment problem, we propose a multi-agent reinforcement learning algorithm with counterfactual reward mechanism, which is termed as CoRe algorithm. CoRe computes the global reward difference in condition that the agent does not take its actual action but takes other actions, while other agents fix their actual actions. This approach can determine each agent's contribution for the global reward. We evaluate CoRe in a simplified Pig Chase game with a decentralised Deep Q Network (DQN) framework. The proposed method helps agents learn end-to-end collaborative behaviors. Compared with other DQN variants with global reward, CoRe significantly improves learning efficiency and achieves better results. In addition, CoRe shows excellent performances in various size game environments.

关键词： Games Learning (artificial intelligence) Machine learning Collaboration Training Task analysis Multi-agent systems

来源：评论

学校读者我要写书评

暂无评论

Shift-Invariant Convolutional Network Search

Shift-Invariant Convolutional Network Search

引用

International Joint Conference on Neural Networks (IJCNN)

作者： Nannan Li Yaran Chen Zixiang Ding Dongbin Zhao State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing China School of artificial intelligence University of Chinese Academy of Sciences Beijing China

ISBN: (数字)9781728169262

ISBN: (纸本)9781728169279

The development of Neural Architecture Search (NAS) makes Convolutional Neural Networks (CNN) more diverse and effective. But previous NAS approaches don't pay attention to the shift-invariant of CNN. Without the shift-invariant, convolutional network is not robust enough when input data is disturbed or damaged. Besides, taking accuracy as the only optimization goal of NAS cannot meet the increasingly diverse needs. In this paper, we propose Shift-Invariant Convolutional Network Search (SICNS). It uses one-shot NAS to search for shift-invariant convolutional network by incorporating the low-pass filter into the one-shot model. Furthermore, SICNS optimizes multiple indicators simultaneously through the multi-objective evolutionary algorithm. Through training one-shot model and evolving the architecture, we obtain convolutional networks which are robust and powerful on image classification task. Especially, our work can achieve 4.52% test error on CIFAR-10 with 0.7M parameters. And in case the input data are disturbed, the accuracy of searched network is 2.96% higher than network without low-pass filter.

关键词： Training Optimization Kernel Convolution Adaptation models Task analysis Search problems

来源：评论

学校读者我要写书评

暂无评论

Evolution and Neural Representation of Mammalian Cooperative Behavior

SSRN

引用

SSRN 2021年

作者： Jiang, Mengping Wang, Miaoyaoxin Shi, Qianqian Wei, Lei Lin, Yongqin Wu, Dingcheng Liu, Boyi Nie, Xiupeng Qiao, Hong Xu, Lin Yang, Tianming Wang, Zuoren Institute of Neuroscience State Key Laboratory of Neuroscience CAS Center for Excellence in Brain Science &Intelligence Technology Chinese Academy of Sciences Shanghai200031 China Key Laboratory of Animal Models and Human Disease Mechanisms and Laboratory of Learning and Memory Kunming Institute of Zoology Chinese Academy of Sciences KunmingYunnan650223 China State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China

Cooperation is common in nature and has been pivotal to the development of human society. However, the details of how and why cooperation evolved remain poorly understood. Cross-species investigation of cooperation may help to elucidate the evolution of cooperative strategies. Thus, we designed a novel cooperative behavioral paradigm and quantitatively examined the cooperative abilities and strategies of mice, rats, and tree shrews. We found that social communication played a key role in the establishment of new cooperation, and that increased cooperative ability and a more efficient cooperative strategy emerged as a function of the evolutionary hierarchy of the tested species. We also found neural representations of distinct aspects of cooperation in the brains of rats through in vivo electrophysiological recording. Cooperation-preferential signals were identified in addition to material reward-preferential signals, each of which may represent internal drives for cooperation. © 2021, The Authors. All rights reserved.

关键词： Rats

来源：评论

学校读者我要写书评

暂无评论

ABCP: Automatic block-wise and channel-wise network pruning via joint search

arXiv

引用

arXiv 2021年

作者： Li, Jiaqi Li, Haoran Chen, Yaran Ding, Zixiang Li, Nannan Ma, Mingjun Duan, Zicheng Zhao, Dongbing The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The School of Mechanical Engineering Beijing Institute of Technology Beijing100081 China The University of Chinese Academy of Sciences Beijing China The College of Engineering and Computer Science Australian National University ACT2601 Australia

Currently, an increasing number of model pruning methods are proposed to resolve the contradictions between the computer powers required by the deep learning models and the resource-constrained devices. However, most of the traditional rule-based network pruning methods can not reach a sufficient compression ratio with low accuracy loss and are time-consuming as well as laborious. In this paper, we propose Automatic Block-wise and Channel-wise Network Pruning (ABCP1) to jointly search the block-wise and channel-wise pruning action with deep reinforcement learning. A joint sample algorithm is proposed to simultaneously generate the pruning choice of each residual block and the channel pruning ratio of each convolutional layer from the discrete and continuous search space respectively. The best pruning action taking both the accuracy and the complexity of the model into account is obtained finally. Compared with the traditional rule-based pruning method, this pipeline saves human labor and achieves a higher compression ratio with lower accuracy loss. Tested on the mobile robot detection dataset, the pruned YOLOv3 model saves 99.5% FLOPs, reduces 99.5% parameters, and achieves 37.3× speed up with only 2.8% mAP loss. The results of the transfer task on the sim2real detection dataset also show that our pruned model has much better robustness performance. Copyright © 2021, The Authors. All rights reserved.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Weakly Aligned Feature Fusion for Multi-modal Object Detection

arXiv

引用

arXiv 2022年

作者： Zhang, Lu Liu, Zhiyong Zhu, Xiangyu Song, Zhan Yang, Xu Lei, Zhen Qiao, Hong The State Key Laboratory of Management and Control for Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China University of Chinese Academy of Sciences Beijing100086 China Center for Excellence in Brain Science and Intelligence Technology Chinese Academy of Sciences Shanghai200031 China The Center for Biometrics and Security Research National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing100190 China The Centre for Artificial Intelligence and Robotics Hong Kong Institute of Science & Innovation Chinese Academy of Sciences Hong Kong The Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Guangdong Shenzhen518055 China

To achieve accurate and robust object detection in the real-world scenario, various forms of images are incorporated, such as color, thermal, depth, etc. However, multi-modal data often suffer from the position shift problem, i.e. the image pair is not strictly aligned, making one object has different positions in different modalities. For the deep learning method, this problem makes it difficult to fuse multi-modal features and puzzles the CNN training. In this paper, we propose a general multi-modal detector named Aligned Region CNN (AR-CNN) to tackle the position shift problem. Firstly, a region feature alignment module with adjacent similarity constraint is designed to consistently predict the position shift between two modalities and adaptively align the cross-modal region features. Secondly, we propose a novel RoI jitter strategy to improve the robustness to unexpected shift patterns. Thirdly, we present a new multi-modal feature fusion method, which selects the more reliable feature and suppresses the less useful one via feature re-weighting. Additionally, by locating bounding boxes in both modalities and building their relationships, we provide novel multi-modal labelling named KAIST-Paired. Extensive experiments on 2D and 3D object detection, RGB-T and RGB-D datasets demonstrate the effectiveness and robustness of our method. Copyright © 2022, The Authors. All rights reserved.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：