检索结果-内蒙古大学图书馆

Barrier-certified adaptive reinforcement learning with applications to brushbot navigation

学校读者我要写书评

暂无评论

arXiv 2018年

作者： Ohnishi, Motoya Wang, Li Notomista, Gennaro Egerstedt, Magnus School of Electrical Engineering Royal Institute of Technology Stockholm11428 Sweden Georgia Robotics and Intelligent Systems Laboratory Georgia Institute of Technology AtlantaGA30332 United States RIKEN Center for Advanced Intelligence Project Tokyo103-0027 Japan School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA30332 United States School of Mechanical Engineering Georgia Institute of Technology AtlantaGA30313 United States

This paper presents a safe learning framework that employs an adaptive model learning algorithm together with barrier certificates for systems with possibly nonstationary agent dynamics. To extract the dynamic structure of the model, we use a sparse optimization technique. We use the learned model in combination with control barrier certificates which constrain policies (feedback controllers) in order to maintain safety, which refers to avoiding particular undesirable regions of the state space. Under certain conditions, recovery of safety in the sense of Lyapunov stability after violations of safety due to the nonstationarity is guaranteed. In addition, we reformulate an action-value function approximation to make any kernel-based nonlinear function estimation method applicable to our adaptive learning framework. Lastly, solutions to the barrier-certified policy optimization are guaranteed to be globally optimal, ensuring the greedy policy improvement under mild conditions. The resulting framework is validated via simulations of a quadrotor, which has previously been used under stationarity assumptions in the safe learnings literature, and is then tested on a real robot, the brushbot, whose dynamics is unknown, highly complex and nonstationary. Copyright © 2018, The Authors. All rights reserved.

关键词： Reinforcement learning

Convolutional ordinal regression forest for image ordinal estimation

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Zhu, Haiping Shan, Hongming Zhang, Yuheng Che, Lingfu Xu, Xiaoyang Zhang, Junping Shi, Jianbo Wang, Fei-Yue 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 GRASP Laboratory University of Pennsylvania PhiladelphiaPA United States The State Key Laboratory for Management and Control of Complex Systems Institute of Automation Chinese Academy of Sciences Beijing100190 China The Institute of Systems Engineering Macau University of Science and Technology 999078 China The University of Chinese Academy of Sciences Beijing100049 China

Image ordinal estimation is to predict the ordinal label of a given image, which can be categorized as an ordinal regression problem. Recent methods formulate an ordinal regression problem as a series of binary classification problems. Such methods cannot ensure that the global ordinal relationship is preserved since the relationships among different binary classifiers are neglected. We propose a novel ordinal regression approach, termed Convolutional Ordinal Regression Forest or CORF, for image ordinal estimation, which can integrate ordinal regression and differentiable decision trees with a convolutional neural network for obtaining precise and stable global ordinal relationships. The advantages of the proposed CORF are twofold. First, instead of learning a series of binary classifiers independently, the proposed method aims at learning an ordinal distribution for ordinal regression by optimizing those binary classifiers simultaneously. Second, the differentiable decision trees in the proposed CORF can be trained together with the ordinal distribution in an end-to-end manner. The effectiveness of the proposed CORF is verified on two image ordinal estimation tasks, i.e. facial age estimation and image aesthetic assessment, showing significant improvements and better stability over the state-of-the-art ordinal regression methods. Copyright © 2020, The Authors. All rights reserved.

关键词： Decision trees

Evidence-Aware Multi-Modal Data Fusion and its Application to Total Knee Replacement Prediction

学校读者我要写书评

暂无评论

Evidence-Aware Multi-Modal Data Fusion and its Application t...

Proceedings of the Digital Image Computing: Technqiues and Applications (DICTA)

作者： Xinwen Liu Jing Wang S. Kevin Zhou Craig Engstrom Shekhar S. Chandra School of Electrical Engineering and Computer Science The University of Queensland Brisbane Australia The Commonwealth Scientific and Industrial Research Organisation Canberra Australia Center for Medical Imaging Robotics Analytic Computing & Learning (MIRACLE) School of Biomedical Engineering & Suzhou Institute for Advanced Research University of Science and Technology of China Suzhou China Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Beijing China School of Human Movement and Nutrition Sciences The University of Queensland Brisbane Australia

ISBN: (数字)9798350379037

ISBN: (纸本)9798350379044

Deep neural networks have been widely studied to predict a medical condition, such as total knee replacement (TKR). It has shown that data of different modalities, such as imaging data, clinical variables, and demographic information, provide complementary information and thus can improve the prediction accuracy together. However, the data sources of various modalities may not always be of high quality, and each modality may have only partial information of medical condition. Thus, predictions from different modalities can be in conflict, and the final prediction may fail in the presence of such a conflict. Therefore, it is important to account for the reliability of each source data and the prediction output when making a final decision. In this paper, we propose an evidence-aware multimodal data fusion framework based on the Dempster-Shafer theory (DST). The backbone models contain an image branch, a non-image branch and a fusion branch. For each branch, there is an evidence network that takes the extracted features as input and outputs an evidence score, which is designed to represent the reliability of the output from the current branch. The output probabilities along with the evidence scores from multiple branches are combined with the Dempster's combination rule to make a final prediction. Experimental results on the public OA initiative (OAI) dataset for the TKR prediction task show that the proposed method has better performance by accounting for conflicts from various modalities.

关键词： Medical conditions Soft sensors Digital images Data integration Imaging Artificial neural networks Reliability theory Feature extraction Reliability engineering Knee replacement

EEG-Based Brain-Computer Interfaces Are Vulnerable to Backdoor Attacks

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Meng, Lubin Huang, Jian Zeng, Zhigang Jiang, Xue Yu, Shan Jung, Tzyy-Ping Lin, Chin-Teng Chavarriaga, Ricardo Wu, Dongrui Ministry of Education Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China Brainnetome Center National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing China La Jolla CA United States Center for Advanced Neurological Engineering Institute of Engineering in Medicine Ucsd La Jolla CA United States Centre of Artificial Intelligence Faculty of Engineering and Information Technology University of Technology Sydney Australia Zhaw DataLab Zürich University of Applied Sciences Winterthur8401 Switzerland

Research and development of electroencephalogram (EEG) based brain-computer interfaces (BCIs) have advanced rapidly, partly due to deeper understanding of the brain and wide adoption of sophisticated machine learning approaches for decoding the EEG signals. However, recent studies have shown that machine learning algorithms are vulnerable to adversarial attacks. This article proposes to use narrow period pulse for poisoning attack of EEG-based BCIs, which is implementable in practice and has never been considered before. One can create dangerous backdoors in the machine learning model by injecting poisoning samples into the training set. Test samples with the backdoor key will then be classified into the target class specified by the attacker. What most distinguishes our approach from previous ones is that the backdoor key does not need to be synchronized with the EEG trials, making it very easy to implement. The effectiveness and robustness of the backdoor attack approach is demonstrated, highlighting a critical security concern for EEG-based BCIs and calling for urgent attention to address it. Copyright © 2020, The Authors. All rights reserved.

关键词： Electroencephalography

EEG-Based Brain-Computer Interfaces Are Vulnerable to Backdoor Attacks

学校读者我要写书评

暂无评论

Research Square

Research Square 2021年

作者： Meng, Lubin Huang, Jian Zeng, Zhigang Jiang, Xue Yu, Shan Jung, Tzyy-Ping Lin, Chin-Teng Chavarriaga, Ricardo Wu, Dongrui Ministry of Education Key Laboratory of Image Processing and Intelligent Control School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China Brainnetome Center and National Laboratory of Pattern Recognition Institute of Automation Chinese Academy of Sciences Beijing China La Jolla CA United States Center for Advanced Neurological Engineering Institute of Engineering in Medicine UCSD La Jolla CA United States Centre of Artificial Intelligence Faculty of Engineering and Information Technology University of Technology Sydney Australia ZHAW DataLab Zürich University of Applied Sciences Winterthur8401 Switzerland

Research and development of electroencephalogram (EEG) based brain-computer interfaces (BCIs) have advanced rapidly, partly due to the wide adoption of sophisticated machine learning approaches for decoding the EEG signals. However, recent studies have shown that machine learning algorithms are vulnerable to adversarial attacks, e.g., the attacker can add tiny adversarial perturbations to a test sample to fool the model, or poison the training data to insert a secret backdoor. Previous research has shown that adversarial attacks are also possible for EEG-based BCIs. However, only adversarial perturbations have been considered, and the approaches are theoretically sound but very difficult to implement in practice. This article proposes to use narrow period pulse for poisoning attack of EEG-based BCIs, which is more feasible in practice and has never been considered before. One can create dangerous backdoors in the machine learning model by injecting poisoning samples into the training set. Test samples with the backdoor key will then be classified into the target class specified by the attacker. What most distinguishes our approach from previous ones is that the backdoor key does not need to be synchronized with the EEG trials, making it very easy to implement. The effectiveness and robustness of the backdoor attack approach is demonstrated, highlighting a critical security concern for EEG-based BCIs. © 2021, CC BY.

关键词： Electroencephalography

Hair Direction Detection*

学校读者我要写书评

暂无评论

Hair Direction Detection*

WRC Symposium on advanced robotics and Automation (WRC SARA)

作者： Peng Ba Pengyi Wang Hongde Wu Qian Yang Yongqiang Feng Junchen Wang Yida David Hu Changsheng Li Wenyong Liu Shaolong Kuang Bai-Quan Su Medical Robotics Laboratory School of Intelligent Engineering and Automation Beijing University of Posts and Telecommunications Beijing China Plastic Surgery Hospital Chinese Academy of Medical Sciences and Peking Union Medical College Beijing China School of Mechanical Engineering and Automation Beihang University Beijing China Brigham and Women’s Hospital Harvard Medical School Boston MA USA School of Mechatronical Engineering Beijing Institute of Technology Beijing China Beijing Advanced Innovation Center for Intelligent Robots and Systems Beijing Institute of Technology Beijing China Key Laboratory for Biomechanics and Mechanobiology of Chinese Education Ministry Beijing Advanced Innovation Centre for Biomedical Engineering School of Biological Science and Medical Engineering Beihang University Beijing China School of Health Science and Environmental Engineering Shenzhen University of Technology Shenzhen China

ISBN: (数字)9798331506100

ISBN: (纸本)9798331506117

Hair direction is an important external feature of hair, and recognising hair direction is a prerequisite for processing hair. In this paper, a new algorithm is proposed and systematically verified experimentally for the problem of recognising hair direction. The main goal of this paper is to develop an algorithm that can identify hair direction in a complex image environment. A curve segment analysis method based on image skeletonisation is adopted, which is based on skeleton extraction, intersection identification, curve segmentation and direction prediction. In addition, this paper combines the technique of non-maximal value suppression and PCA analysis to improve the accuracy and stability of the estimation. In the experimental design, this paper chooses a representative image dataset to verify the effectiveness of this paper's algorithm. The experimental process includes the steps of image preprocessing, skeletonisation processing, intersection detection and merging, and direction prediction. The experimental results show that the method in this paper can accurately and effectively identify the hair direction. The main contribution of this paper is to propose a new hair direction recognition method and experimentally verify its effectiveness and accuracy in complex backgrounds.

关键词： Hair Image segmentation Accuracy Three-dimensional displays Merging Prediction algorithms Stability analysis

A Haptic Exploration and Surface Classification of Objects with Four Typical Surface Properties

学校读者我要写书评

暂无评论

A Haptic Exploration and Surface Classification of Objects w...

International Conference on advanced robotics and Mechatronics (ICARM)

作者： Peng Qi Yunfeng Wu Tianliang Yao Bo Lu Yi Sun Jian S. Dai Department of Control Science and Engineering College of Electronics and Information Engineering Tongji University Shanghai P. R. China Robotics and Microsystems Center School of Mechanical and Electric Engineering Soochow University Suzhou Jiangsu China Laboratory of Intelligent Systems Ecole Polytechnique Federale de Lausanne Lausanne Switzerland Department of Mechanical and Energy Engineering College of Engineering Southern University of Science and Technology Shenzhen P. R. China School of Natural Mathematical and Engineering Sciences King's College London

To effectively interact with the physical world, an intelligent robot is required to have the ability to obtain the detailed features of an unknown object. Visual devices are commonly used to detect the global geometry of an object; however, detailed information such as surface properties cannot be identified using these devices. The current study proposes an efficient haptic exploration method to recognize the physical properties of surfaces using an intelligent fingertip. Our surface-following algorithm utilizes the normal force vector and the tangential force vector at the contact point between the fingertip and the target object to predict the moving direction and implement the surface exploration. And a correction index $K$ is introduced to adjust the sliding velocity. The algorithm is proved to be robust and high performance both in simulation and lab experiments. Finally, we propose and explore a haptic prediction neural network, which enables our robot to have an accurate feel through physical interaction. In the classification experiment of four types of objects with different surface properties, the average accuracy of the proposed model is 90.2%.

关键词：

Point spread function modelling for wide field small aperture telescopes with a denoising autoencoder

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Jia, Peng Li, Xiyu Li, Zhengyang Wang, Weinan Cai, Dongmei College of Physics and Optoelectronics Taiyuan University of Technology Taiyuan030024 China Key Laboratory of Advanced Transducers and Intelligent Control Systems Ministry of Education and Shanxi Province Taiyuan University of Technology Taiyuan030024 China Nanjing Institute of Astronomical Optics and Technology CAS Nanjing Jiangsu210042 China Department of Physics Durham University South Road DurhamDH1 3LE United Kingdom Wuxi Internet of Innovation Center Company Limited Wuxi Jiangsu214135 China

The point spread function reflects the state of an optical telescope and it is important for data post-processing methods design. For wide field small aperture telescopes, the point spread function is hard to model, because it is affected by many different effects and has strong temporal and spatial variations. In this paper, we propose to use the denoising autoencoder, a type of deep neural network, to model the point spread function of wide field small aperture telescopes. The denoising autoencoder is a pure data based point spread function modelling method, which uses calibration data from real observations or numerical simulated results as point spread function templates. According to real observation conditions, different levels of random noise or aberrations are added to point spread function templates, making them as realizations of the point spread function, i.e., simulated star images. Then we train the denoising autoencoder with realizations and templates of the point spread function. After training, the denoising autoencoder learns the manifold space of the point spread function and can map any star images obtained by wide field small aperture telescopes directly to its point spread function, which could be used to design data post-processing or optical system alignment methods. Copyright © 2020, The Authors. All rights reserved.

关键词： Optical transfer function

Self-adaptation of chimera states

学校读者我要写书评

暂无评论

Physical Review E 2019年第1期99卷 010201(R)-010201(R)页

作者： Nan Yao Zi-Gang Huang Hai-Peng Ren Celso Grebogi Ying-Cheng Lai Department of Applied Physics Xi'an University of Technology Xi'an 710048 China The Key Laboratory of Biomedical Information Engineering of Ministry of Education National Engineering Research Center of Health Care and Medical Devices The Key Laboratory of Neuro-informatics & Rehabilitation Engineering of Ministry of Civil Affairs and Institute of Health and Rehabilitation Science School of Life Science and Technology Xi'an Jiaotong University Xi'an 710049 China Shaanxi Key Laboratory of Complex System Control and Intelligent Information Processing Xi'an University of Technology Xi'an 710048 China Institute for Complex Systems and Mathematical Biology King's College University of Aberdeen Aberdeen AB24 3UE United Kingdom School of Electrical Computer and Energy Engineering Arizona State University Tempe Arizona 85287 USA Department of Physics Arizona State University Tempe Arizona 85287 USA

Chimera states in spatiotemporal dynamical systems have been investigated in physical, chemical, and biological systems, and have been shown to be robust against random perturbations. How do chimera states achieve their robustness? We uncover a self-adaptation behavior by which, upon a spatially localized perturbation, the coherent component of the chimera state spontaneously drifts to an optimal location as far away from the perturbation as possible, exposing only its incoherent component to the perturbation to minimize the disturbance. A systematic numerical analysis of the evolution of the spatiotemporal pattern of the chimera state towards the optimal stable state reveals an exponential relaxation process independent of the spatial location of the perturbation, implying that its effects can be modeled as restoring and damping forces in a mechanical system and enabling the articulation of a phenomenological model. Not only is the model able to reproduce the numerical results, it can also predict the trajectory of drifting. Our finding is striking as it reveals that, inherently, chimera states possess a kind of “intelligence” in achieving robustness through self-adaptation. The behavior can be exploited for the controlled generation of chimera states with their coherent component placed in any desired spatial region of the system.

关键词： Chimera states Collective dynamics Coupled oscillators Chaos & nonlinear dynamics