检索结果-内蒙古大学图书馆

Proceedings of the 2019 2nd International Conference on Signal Processing and Machine Learning

作者： Enwei Zhang Hongtu Zhou Yongchao Ding Junqiao Zhao Chen Ye The Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University School of Electronics and Information Engineering Shanghai P. R. China Tongji University School of Electronics and Information Engineering Shanghai P. R. China Tongji University School of Electronics and Information Engineering P. R. China The Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University School of Electronics and Information Engineering P. R. China

ISBN: (纸本)9781450372213

Obstacle avoiding is one of the most complex tasks for autonomous driving systems, which was also ignored by many cutting-edge end-to-end learning-based methods. The difficulties stem from the integrated process of detection and interpretation of environment and obstacles and generation of proper behaviors. We make the use of CARLA, a simulator for autonomous driving research, and collect massive human drivers' reactions to obstacles on road subjecting to given driving commands, i.e. follow, go straight, turn left and turn right for about 6 hours. A behavior-Cloning neural network architecture is proposed with the modified loss that enlarge the effects of errors for steer, which indicates the benefit to high an accuracy. We found the data augmentation of the image is crucial to the training of the proposed network. And a reasonable limit allows avoiding unexpected stop. The experiments demonstrate 3 obstacle avoidance cases: for the same type as the training dataset, other automobile and two-wheeled vehicles. Finally, the CARLA benchmark is also tested.

关键词： imitation learning obstacle avoidance end-to-end autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

A New Class of Learning Automata for Selecting an Optimal Subset

A New Class of Learning Automata for Selecting an Optimal Su...

引用

IEEE International Conference on systems, Man, and Cybernetics

作者： JunQi Zhang ZeZhou Li Qi Kang MengChu Zhou Key Laboratory of Embedded System and Service Computing Ministry of Education Department of Computer Science and Technology Tongji University Department of Control Science and Engineering Tongji University Department of Electrical and Computer Engineering New Jersy Institute of Technology

ISBN: (纸本)9781479938414

Interacting with a random environment, Learning Automata (LAs) are automata that, generally, have the task of learning the optimal action based on responses from the environment. Distinct from the traditional goal of Learning Automata to select only the optimal action out of a set of actions, this paper considers a multiple-action selection problem and proposes a novel class of Learning Automata for selecting an optimal subset of actions. Their objective is to identify the optimal subset: the top k out of r actions. Based on conventional continuous pursuit and discretized pursuit learning schemes, this paper introduces four pursuit learning schemes for selecting the optimal subset, called continuous equal pursuit, discretized equal pursuit, continuous unequal pursuit and discretized unequal pursuit learning schemes, respectively. In conjunction with a reward-inaction learning paradigm, the above four schemes lead to four versions of pursuit Learning Automata for selecting the optimal subset. The simulation results present a quantitative comparison between them.

关键词： Learning Automata Optimal subset Pursuit algorithm

来源：评论

学校读者我要写书评

暂无评论

Object rigidity and reflectivity identification based on motion analysis

Object rigidity and reflectivity identification based on mot...

引用

IEEE International Conference on Image Processing

作者： Di Zang Paul R. Schrater Katja Doerschner Department of Computer Science and Engineering University of Minnesota USA Key Laboratory of Embedded System and Service Computing Ministry of Education University of Tongji China National Research Center of Magnetic Resonance and Department of Psychology Bilkent University Turkey

Rigidity and reflectivity are important properties of objects, identifying these properties is a fundamental problem for many computer vision applications like motion and tracking. In this paper, we extend our previous work to propose a motion analysis based approach for detecting the object's rigidity and reflectivity. This approach consists of two steps. The first step aims to identify object rigidity based on motion estimation and optic flow matching. The second step is to classify specular rigid and diffuse rigid objects using structure from motion and Procrustes analysis. We show how rigid bodies can be detected without knowing any prior motion information by using a mutual information based matching method. In addition, we use a statistic way to set thresholds for rigidity classification. Presented results demonstrate that our approach can efficiently classify the rigidity and reflectivity of an object.

关键词： Optical imaging Adaptive optics Mutual information Shape Computer vision Integrated optics Object recognition

来源：评论

学校读者我要写书评

暂无评论

Structure-Aware Generative Adversarial Network for Text-to-Image Generation

Structure-Aware Generative Adversarial Network for Text-to-I...

引用

IEEE International Conference on Image Processing

作者： Wenjie Chen Zhangkai Ni Hanli Wang Department of Computer Science and Technology Tongji University Shanghai P. R. China Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai P. R. China

Text-to-image generation aims at synthesizing photo-realistic images from textual descriptions. Existing methods typically align images with the corresponding texts in a joint semantic space. However, the presence of the modality gap in the joint semantic space leads to misalignment. Meanwhile, the limited receptive field of the convolutional neural network leads to structural distortions of generated images. In this work, a structure-aware generative adversarial network (SaGAN) is proposed for (1) semantically aligning multimodal features in the joint semantic space in a learnable manner; and (2) improving the structure and contour of generated images by the designed content-invariant negative samples. Experimental results show that SaGAN achieves over 30.1% and 8.2% improvements in terms of FID on the datasets of CUB and COCO when compared with the state-of-the-art approaches.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improved asymptotic multicast throughput for random extended networks

引用

Computer Communications 2010年第18期33卷 2195-2202页

作者： Cheng Wang Changjun Jiang Xiang-Yang Li Jiujun Cheng Department of Computer Science and Technology Tongji University Shanghai China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications China Key Laboratory of Embedded System and Service Computing Ministry of Education Shanghai China Department of Computer Science Illinois Institute of Technology Chicago IL 60616 USA

We study the asymptotic throughput for random extended networks , where n ad hoc nodes are randomly deployed in a square region R ( n ) = 0 , n 2 . We directly consider the multicast throughput to unify the unicast and broadcast throughput, and design a new multicast scheme under the generalized physical model based on the so-called secondary highways system . Taking account of all possible cases of n s = ω (1) and 1 ⩽ n d ⩽ n − 1, we derive the achievable multicast throughput, where n s and n d denote the number of sessions and the number of destinations of each session. We prove that for some cases in terms of n s and n d , our scheme achieves better throughput than the existing schemes.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Petri net modeling and one-wafer scheduling of single-arm multi-cluster tools

Petri net modeling and one-wafer scheduling of single-arm mu...

引用

2013 IEEE International Conference on Automation Science and Engineering, CASE 2013

作者： Zhu, Qinghua Wu, Naiqi Qiao, Yan Zhou, Mengchu School of Computer Guangdong University of Technology Guangzhou 510006 China Macau University of Science and Technology China Department of Industrial Engineering School of Electro-Mechanical Engineering Guangdong University of Technology Guangzhou 510006 China Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai 201804 China Department of Electrical and Computer Engineering New Jersey Institute of Technology Newark NJ 07102-1982 United States

ISBN: (纸本)9781479915156

It is very challenging to schedule a multi-cluster tool for it needs to coordinate the activities of multiple robots. This paper studies one-wafer cyclic scheduling for multi-cluster tools whose bottleneck cluster tool is process-bound. A Petri net model is developed to describe the system. Based on the model, it is shown that, for any multi-cluster tool whose bottleneck cluster tool is process-bound, there is always a one-wafer cyclic schedule. Then, a method is presented to find the minimal cycle time and the optimal one-wafer cyclic schedule. The method requires only simple calculation. An example illustrates the application and effectiveness of the proposed method. © 2013 IEEE.

关键词： Scheduling

来源：评论

学校读者我要写书评

暂无评论

Knowledge Granulation in Interval-Valued Information systems

Knowledge Granulation in Interval-Valued Information Systems

引用

International Workshop on Intelligent systems and Applications, ISA

作者： Nan Zhang Duoqian Miao Can Gao Department of Computer Science and Technology University of Tongji Shanghai China Department of Electrical and Computer Engineering University of Alberta Edmonton AB Canada Key Laboratory of Embedded System and Service Computing of Ministry of Education University of Tongji Shanghai China

Rough set theory, proposed by Pawlak, has been proved to be a mathematical tool to deal with vagueness and uncertainty in intelligent information processing. In this paper, we propose the concept of knowledge granulation in interval-valued information systems, and discuss some important properties. From these properties, it can be shown that the proposed knowledge granulation provides important approaches to measuring the discernibility of different knowledge. It may be helpful for rule evaluation and knowledge discovery in interval-valued information systems.

关键词： Information systems Approximation methods Uncertainty Rough sets Computers Entropy

来源：评论

学校读者我要写书评

暂无评论

On Theoretical Optimization of the Sensing Matrix for Sparse-Dictionary Signal Recovery

On Theoretical Optimization of the Sensing Matrix for Sparse...

引用

IEEE Global Conference on Signal and Information Processing (GlobalSIP)

作者： Jianchen Zhu Shengjie Zhao Xu Ma Gonzalo R. Arce Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai CHN School of Optics and Photonics Beijing Institute of Technology Beijing CHN Department of Electrical and Computer Engineering University of Delaware Newark Delaware USA

Compressive Sensing (CS) is a new paradigm for the efficient acquisition of signals that have sparse representation in a certain domain. Traditionally, CS has provided numerous methods for signal recovery over an orthonormal basis. However, modern applications have sparked the emergence of related methods for signals not sparse in an orthonormal basis but in some arbitrary, perhaps highly overcomplete, dictionary, particularly due to their potential to generate different kinds of sparse representation of signals. Here, we first propose the Signal space Subspace Pursuit (SSSP) algorithm, and then we derive a low bound on the number of measurements required. The algorithm has low computational complexity and provides high recovery accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

VCFGAN: A Two-stage Collaborative Filtering Framework based on Variational Autoencoders and Generative Adversarial Networks

VCFGAN: A Two-stage Collaborative Filtering Framework based ...

引用

Information systems and computing Technology (ISCTech), International Conference on

作者： Zhiqiang Huang Jinzhong Li Xueyu Huang Junqi Liu School of Software Engineering Jiangxi University of Science and Technology Nanchang China College of Electronics and Information Engineering Jinggangshan University Ji’an China Ministry of Education Key Laboratory of Embedded System and Service Computing (Tongji University) Shanghai China

ISBN: (数字)9798350379860

ISBN: (纸本)9798350379877

Deep generative models, such as Generative Adversarial Networks (GAN) and Variational Autoencoders (VAE), are widely used in collaborative filtering. They usually learn users’ preferences for items directly from a highly sparse user rating matrix (URM), and then recommend the top-N items to users. Due to the high sparsity of URM, GAN has limited ability to handle sparse data, while VAE’s encoder can extract features. Therefore, we propose a two-stage collaborative filtering framework based on variational autoencoders and generative adversarial networks, named VCFGAN. It first uses two VAEs to extract features from URM and side information (SI), and then uses the extracted latent vectors to train the GAN network. To evaluate the performance of our proposed VCFGAN model, some experiments are conducted on two real datasets, and the experimental results show that our model outperforms other representative models.

关键词： Training Collaborative filtering Autoencoders Generative adversarial networks Feature extraction Vectors Data mining Sparse matrices Recommender systems Information systems

来源：评论

学校读者我要写书评

暂无评论

Berth-Crane allocation model and algorithm of container berths in the uncertain environment

Advances in Intelligent Systems and Computing

引用

Advances in Intelligent systems and computing 2013年 212卷 785-792页

作者： Liu, Jian-xin Du, Yu-yue Hong, Yong-fa Sun, Lin-lin College of Information Science Engineering Shandong University of Science and Technology Qingdao 266590 Shandong China The Key Laboratory of Embedded System and Service Computing Ministry of Education Tongji University Shanghai 200092 China College of Civil Engineering and Architecture SUST Qingdao 266590 Shandong China

ISBN: (纸本)9783642375019

According to the randomness of the vessel's arrival time and handling time, the establishment of a randomly-oriented environment container berths-crane allocation model, the optimizing goal is to minimize the average ship-waiting time. Taking into account the complexity of the model solution, this paper offers the design of an improved genetic algorithm to reduce the searching space, and according to the characteristics of the optimal solution. With testing example to verify that the model can simulate decision-making environment of berths-crane allocation problem and reflect the decision-maker's attitude toward risks and preferences. The algorithm can gain a stable and satisfactory solution within the operating time. © Springer-Verlag Berlin Heidelberg 2013.

关键词： Genetic algorithms

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：