DOCSLIB.ORG

Transfer Learning in Deep Reinforcement Learning: a Survey

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Transfer Learning in Deep Reinforcement Learning: a Survey 1 Transfer Learning in Deep Reinforcement Learning: A Survey Zhuangdi Zhu, Kaixiang Lin, and Jiayu Zhou Abstract—Reinforcement Learning (RL) is a key technique to address sequential decision-making problems and is crucial to realize advanced artificial intelligence. Recent years have witnessed remarkable progress in RL by virtue of the fast development of deep neural networks. Along with the promising prospects of RL in numerous domains, such as robotics and game-playing, transfer learning has arisen as an important technique to tackle various challenges faced by RL, by transferring knowledge from external expertise to accelerate the learning process. In this survey, we systematically investigate the recent progress of transfer learning approaches in the context of deep reinforcement learning. Specifically, we provide a framework for categorizing the state-of-the-art transfer learning approaches, under which we analyze their goals, methodologies, compatible RL backbones, and practical applications. We also draw connections between transfer learning and other relevant topics from the RL perspective and explore their potential challenges as well as open questions that await future research progress. Index Terms—Transfer Learning, Reinforcement Learning, Deep Learning, Survey. F 1 INTRODUCTION Einforcement Learning (RL) is an effective framework even incur safety concerns for domains such as automatic- R to solve sequential decision-making tasks, where a driving and health informatics, where the consequences of learning agent interacts with the environment to improve wrong decisions can be too high to take. The abovementioned its performance through trial and error [1]. Originated challenges have motivated various efforts to improve the from cybernetics and thriving in Computer Science, RL has current RL procedure. As a result, Transfer Learning (TL), been widely applied to tackle challenging tasks which were which is a technique to utilize external expertise from other previously intractable [2, 3]. tasks to benefit the learning process of the target task, As a pioneering technique for realizing advanced arti- becomes a crucial topic in RL. ficial intelligence, traditional RL was mostly designed for TL techniques have been extensively studied in the super- tabular cases, which provided principled solutions to simple vised learning domain [12], whereas it is an emerging topic tasks but faced difficulties when handling highly complex in RL. In fact, TL under the framework of RL can be more domains, e.g. tasks with 3D environments. Over the recent complicated in that the knowledge needs to transfer in the years, an integrated framework, where an RL agent is built context of a Markov Decision Process (MDP). Moreover, due upon deep neural networks, has been developed to address to the delicate components of an MDP, the expert knowledge more challenging tasks. The combination of deep learning may take different forms, which need to transfer in different with RL is hence referred to as Deep Reinforcement Learning ways. Noticing that previous efforts on summarizing TL for (DRL) [4], which aims to address complex domains that RL have not covered its most recent advancement [13, 14], were otherwise unresolvable by building deep, powerful in this survey, we make a comprehensive investigation function approximators. DRL has achieved notable success of Transfer Learning in Deep Reinforcement Learning. Es- in applications such as robotics control [5, 6] and game pecially, we build a systematic framework to categorize arXiv:2009.07888v4 [cs.LG] 4 Mar 2021 playing [7]. It also has a promising prospects in domains such the state-of-the-art TL techniques into different sub-topics, as health informatics [8], electricity networks [9], intelligent review their theories and applications, and analyze their transportation systems[10, 11], to name just a few. inter-connections. Besides its remarkable advancement, RL still faces in- The rest of this survey is organized as follows: In section 2, triguing difficulties induced by the exploration-exploitation we introduce the preliminaries of RL and its key algorithms, dilemma [1]. Specifically, for practical RL, the environment including those recently designed based on deep neural dynamics are usually unknown, and the agent cannot exploit networks. Next, we clarify the definition of TL in the context its knowledge about the environment to improve its perfor- of RL, and discuss its relevant research topics (Section mance until enough interaction experiences are collected via 2.4). In Section 3, we provide a framework to categorize exploration. Due to partial observability, sparse feedbacks, TL approaches from multiple perspectives, analyze their and the high-dimension in state and action spaces, acquiring fundamental differences, and summarize their evaluation sufficient interaction samples can be prohibitive, which may metrics (Section 3.3). In Section 4, we elaborate on different TL approaches in the context of DRL, organized by the format • Zhuangdi Zhu and Jiayu Zhou are with the Department of Computer of transferred knowledge, such as reward shaping (Section Science and Engineering, Michigan State University, East Lansing, MI, 4.1), learning from demonstrations (Section 4.2), or learning 48823. E-mail: [email protected], [email protected] from teacher policies (Section 4.3). We also investigate TL • Kaixiang Lin is with the Amazon Alexa AI. E-mail: [email protected] approaches by the way that knowledge transfer occurs, such 2 as inter-task mapping (Section 4.4), or learning transferrable expected rewards that an agent can get from s, given that representations (Section 4.5), etc. We discuss the recent the agent follows policy π in the environment M afterward. applications of TL in the context of DRL in Section 5 and Similar to the value-function, each policy also carries a Q- provide some future perspectives and open questions in function, which is defined over the state-action space to Section 6. estimate the quality of taking action a from state s: π E 0 π 0 QM(s; a) = s0∼T (·|s;a) [R(s; a; s ) + γVM(s )] : 2 DEEP REINFORCEMENT LEARNING AND TRANS- FER EARNING The objective for an RL agent is to learn an optimal L ∗ policy πM to maximize the expectation of accumulated In this section, we provide a brief overview of the recent ∗ ∗ rewards, so that: 8s 2 S; πM(s) = arg max QM(s; a); development in RL and the definitions of some key ter- a2A ∗ π minologies. Next, we provide categorizations to organize where QM(s; a) = sup QM(s; a). π different TL approaches, then point out some of the other topics in the context of RL, which are relevant to TL but will not be elaborated in this survey. 2.2 Reinforcement Learning Algorithms In this section, we review the key RL algorithms developed Remark 1. Without losing clarify, for the rest of this survey, we over the recent years, which provide cornerstones for the TL refer to MDPs, domains, and tasks equivalently. approaches discussed in this survey. Prediction and Control: any RL problem can be disas- 2.1 Reinforcement Learning Preliminaries sembled into two subtasks: prediction and control [1]. In the A typical RL problem can be considered as training an prediction phase, the quality of the current policy is being agent to interact with an environment that follows a Markov evaluated. In the control phase, which is also referred to as Decision Process (MPD) [15]. For each interaction with the the policy improvement phase, the learning policy is adjusted MDP, the agent starts with an initial state and performs an based on evaluation results from the prediction step. Policies action accordingly, which yields a reward to guide the agent can be improved by iteratively conducting these two steps, actions. Once the action is taken, the MDP transits to the which is therefore called policy iteration. next state by following the underlying transition dynamics of Policy iterations can be model-free, which means that the the MDP. The agent accumulates the time-discounted rewards target policy is optimized without requiring knowledge along with its interactions with the MDP. A subsequence of the MDP transition dynamics. Traditional model-free of interactions is referred to as an episode. For MDPs with RL includes Monte-Carlo methods, which uses samples of infinite horizons, one can assume that there are absorbing episodes to estimate the value of each state based on complete states, such that any action taken upon an absorbing state episodes starting from that state. Monte-Carlo methods can will only lead to itself and yields zero rewards. All above- be on-policy if the samples are collected by following the mentioned components in the MDP can be represented using target policy, or off-policy if the episodic samples are collected a tuple M = (µ0; S; A; T ; γ; R; S0), in which: by following a behavior policy that is different from the target policy. • µ0 is the set of initial states. Temporal-Difference Learning, or TD-learning for short, • S is the state space. is an alternative to Monte-Carlo for solving the prediction • A is the action space. problem. The key idea behind TD-learning is to learn the • T : S × A × S ! R is the transition probability 0 state quality function by bootstrapping. It can also be extended distribution, where T (s js; a) specifies the probability 0 to solve the control problem so that both value function and of the state transitioning to s upon taking action a policy can get improved simultaneously. TD-learning is one from state s. of the most widely used RL paradigms due to its simplicity • R : S × A × S ! R is the reward distribution, where 0 and general applicability. Examples of on-policy TD-learning R(s; a; s ) is the reward an agent can get by taking 0 algorithms include SARSA [16], Expected SARSA [17], Actor- action a from state s with the next state being s .
Load more

Recommended publications
  • Transfer Adaptation Learning: a Decade Survey
    JOURNAL OF LATEX CLASS FILES, VOL. -, NO. -, MARCH 2019 1 Transfer Adaptation Learning: A Decade Survey Lei Zhang, Xinbo Gao Abstract—The world we see is ever-changing and it always changes with people, things, and the environment. Domain is referred to as the state of the world at a certain moment. A research problem is characterized as transfer adaptation learning (TAL) when it needs knowledge correspondence between different moments/domains. Conventional machine learning aims to find a model with the minimum expected risk on test data by minimizing the regularized empirical risk on the training data, which, however, supposes that the training and test data share similar joint probability distribution. TAL aims to build models that can perform tasks of target domain by learning knowledge from a semantic related but distribution different source domain. It is an energetic research filed of increasing influence and importance, which is presenting a blowout publication trend. This paper surveys the advances of TAL methodologies in the past decade, and the technical challenges and essential problems of TAL have been observed and discussed with deep insights and new perspectives. Broader solutions of transfer adaptation learning being created by researchers are identified, i.e., instance re-weighting adaptation, feature adaptation, classifier adaptation, deep network adaptation and adversarial adaptation, which are beyond the early semi-supervised and unsupervised split. The survey helps researchers rapidly but comprehensively understand and identify the research foundation, research status, theoretical limitations, future challenges and under-studied issues (universality, interpretability, and credibility) to be broken in the field toward universal representation and safe applications in open-world scenarios.
    [Show full text]
  • Deep Learning for Electromyographic Hand Gesture Signal Classification
    1 Deep Learning for Electromyographic Hand Gesture Signal Classification Using Transfer Learning Ulysse Cotˆ e-Allard,´ Cheikh Latyr Fall, Alexandre Drouin, Alexandre Campeau-Lecours, Clement´ Gosselin, Kyrre Glette, Franc¸ois Laviolettey, and Benoit Gosseliny Abstract—In recent years, deep learning algorithms have widely adopted both in research and clinical settings. The become increasingly more prominent for their unparalleled sEMG signals, which are non-stationary, represent the sum ability to automatically learn discriminant features from large of subcutaneous motor action potentials generated through amounts of data. However, within the field of electromyography- based gesture recognition, deep learning algorithms are seldom muscular contraction [1]. Artificial intelligence can then be employed as they require an unreasonable amount of effort from leveraged as the bridge between sEMG signals and the pros- a single person, to generate tens of thousands of examples. thetic behavior. This work’s hypothesis is that general, informative features can The literature on sEMG-based gesture recognition primarily be learned from the large amounts of data generated by aggre- focuses on feature engineering, with the goal of characterizing gating the signals of multiple users, thus reducing the recording burden while enhancing gesture recognition. Consequently, this sEMG signals in a discriminative way [1], [2], [3]. Recently, paper proposes applying transfer learning on aggregated data researchers have proposed deep learning approaches [4], [5], from multiple users, while leveraging the capacity of deep learn- [6], shifting the paradigm from feature engineering to feature ing algorithms to learn discriminant features from large datasets. learning. Regardless of the method employed, the end-goal Two datasets comprised of 19 and 17 able-bodied participants remains the improvement of the classifier’s robustness.
    [Show full text]
  • A Comprehensive Survey on Transfer Learning
    1 A Comprehensive Survey on Transfer Learning Fuzhen Zhuang, Zhiyuan Qi, Keyu Duan, Dongbo Xi, Yongchun Zhu, Hengshu Zhu, Senior Member, IEEE, Hui Xiong, Fellow, IEEE, and Qing He Abstract—Transfer learning aims at improving the performance of target learners on target domains by transferring the knowledge contained in different but related source domains. In this way, the dependence on a large number of target domain data can be reduced for constructing target learners. Due to the wide application prospects, transfer learning has become a popular and promising area in machine learning. Although there are already some valuable and impressive surveys on transfer learning, these surveys introduce approaches in a relatively isolated way and lack the recent advances in transfer learning. Due to the rapid expansion of the transfer learning area, it is both necessary and challenging to comprehensively review the relevant studies. This survey attempts to connect and systematize the existing transfer learning researches, as well as to summarize and interpret the mechanisms and the strategies of transfer learning in a comprehensive way, which may help readers have a better understanding of the current research status and ideas. Unlike previous surveys, this survey paper reviews more than forty representative transfer learning approaches, especially homogeneous transfer learning approaches, from the perspectives of data and model. The applications of transfer learning are also briefly introduced. In order to show the performance of different transfer learning models, over twenty representative transfer learning models are used for experiments. The models are performed on three different datasets, i.e., Amazon Reviews, Reuters-21578, and Office-31.
    [Show full text]
  • Personalizing EEG-Based Affective Models with Transfer Learning
    Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence (IJCAI-16) Personalizing EEG-Based Affective Models with Transfer Learning 1 1,2,3, Wei-Long Zheng and Bao-Liang Lu ⇤ 1Center for Brain-like Computing and Machine Intelligence Department of Computer Science and Engineering 2Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineering 3Brain Science and Technology Research Center Shanghai Jiao Tong University, Shanghai, China weilong, bllu @sjtu.edu.cn { } Abstract way to enhance current BCI systems with an increase of infor- mation flow, while at the same time without additional cost. Individual differences across subjects and non- Therefore, aBCIs have attracted increasing interests in both stationary characteristic of electroencephalography research and industry communities, and various studies have (EEG) limit the generalization of affective brain- presented their efficiency and feasibility [Muhl¨ et al., 2014a; computer interfaces in real-world applications. On 2014b; Jenke et al., 2014; Eaton et al., 2015]. Although the the other hand, it is very time consuming and large progress has been obtained about development of aB- expensive to acquire a large number of subject- CIs in recent years, there still exist some challenges such as specific labeled data for learning subject-specific the adaptation to changing environments and individual dif- models. In this paper, we propose to build per- ferences. sonalized EEG-based affective models without la- beled target data using transfer learning techniques. Until now, most of studies emphasized choices of fea- We mainly explore two types of subject-to-subject tures and classifiers [Singh et al., 2007; Jenke et al., 2014; transfer approaches.
    [Show full text]
  • Transfer Learning: Introduction & Application
    Transfer Learning: Introduction & Application Yun Gu Department of Automation Shanghai Jiao Tong University Shanghai, CHINA June 16th, 2014 Overview of Transfer Learning Categorization Applications Conclusion Outline 1 Overview of Transfer Learning 2 Categorization Three Research Issues Different Settings 3 Applications Image Annotation Image Classification Deep Learning 4 Conclusion Y. Gu Transfer Learning: Introduction & Application Overview of Transfer Learning Categorization Applications Conclusion Outline of this section 1 Overview of Transfer Learning 2 Categorization 3 Applications 4 Conclusion Y. Gu Transfer Learning: Introduction & Application Transfer Learning · What to “Transfer”? · Machiine Learniing Scheme.. · How to “Transfer”? · Rellatiionshiip wiith other ML · When to “Transfer”? tech? In Top-Level Conference: 7 papers in CVPR 2014 are related with Transfer Learning; In Top-Level Journal: M.Guilaumin, et.al, "ImageNet Auto-Annotation with Segmentation Propagation", IJCV,2014 Overview of Transfer Learning Categorization Applications Conclusion What is Transfer Learning?[Pan and Yang, 2010] Naive View (Transfer Learning) Transfer Learning (i.e. Knowledge Transfer, Domain Adaption) aims at applying knowledge learned previously to solve new problems faster or with better solutions. Y. Gu Transfer Learning: Introduction & Application Transfer Learning · What to “Transfer”? · Machiine Learniing Scheme.. · How to “Transfer”? · Rellatiionshiip wiith other ML · When to “Transfer”? tech? In Top-Level Conference: 7 papers in CVPR 2014 are related with Transfer Learning; In Top-Level Journal: M.Guilaumin, et.al, "ImageNet Auto-Annotation with Segmentation Propagation", IJCV,2014 Overview of Transfer Learning Categorization Applications Conclusion What is Transfer Learning?[Pan and Yang, 2010] Naive View (Transfer Learning) Transfer Learning (i.e. Knowledge Transfer, Domain Adaption) aims at applying knowledge learned previously to solve new problems faster or with better solutions.
    [Show full text]
  • Transfer Learning for Reinforcement Learning Domains: a Survey
    JournalofMachineLearningResearch10(2009)1633-1685 Submitted 6/08; Revised 5/09; Published 7/09 Transfer Learning for Reinforcement Learning Domains: A Survey Matthew E. Taylor∗ [email protected] Computer Science Department The University of Southern California Los Angeles, CA 90089-0781 Peter Stone [email protected] Department of Computer Sciences The University of Texas at Austin Austin, Texas 78712-1188 Editor: Sridhar Mahadevan Abstract The reinforcement learning paradigm is a popular way to address problems that have only limited environmental feedback, rather than correctly labeled examples, as is common in other machine learning contexts. While significant progress has been made to improve learning in a single task, the idea of transfer learning has only recently been applied to reinforcement learning tasks. The core idea of transfer is that experience gained in learning to perform one task can help improve learning performance in a related, but different, task. In this article we present a framework that classifies transfer learning methods in terms of their capabilities and goals, and then use it to survey the existing literature, as well as to suggest future directions for transfer learning work. Keywords: transfer learning, reinforcement learning, multi-task learning 1. Transfer Learning Objectives In reinforcement learning (RL) (Sutton and Barto, 1998) problems, leaning agents take sequential actions with the goal of maximizing a reward signal, which may be time-delayed. For example, an agent could learn to play a game by being told whether it wins or loses, but is never given the “correct” action at any given point in time. The RL framework has gained popularity as learning methods have been developed that are capable of handling increasingly complex problems.
    [Show full text]
  • Transfer Learning Approach for Occupancy Prediction in Smart Buildings
    Transfer Learning Approach for Occupancy Prediction in Smart Buildings Mohamad Khalil Stephen McGough Zoya Pourmirza School of Engineering School of Computing School of Engineering Newcastle University Newcastle University Newcastle University Newcastle, Uk Newcastle, UK Newcastle, UK [email protected] [email protected] [email protected] Mehdi Pazhoohesh Sara Walker Faculty of Technology School of Engineering De Montfort University Newcastle University Leicester, UK Newcastle, UK [email protected] [email protected] Abstract— Accurate occupancy prediction in smart I. INTRODUCTION buildings is a key element to reduce building energy consumption and control HVAC systems (Heating – Ventilation In buildings the term of occupancy information refers to and– Air Conditioning) efficiently, resulting in an increment of occupants’ presence or absence and their movements. human comfort. This work focuses on the problem of occupancy Accurate real-time occupancy information in smart prediction modelling (occupied / unoccupied) in smart buildings buildings can help HVAC (Heating – Ventilation and– Air using environmental sensor data. A novel transfer learning Conditioning) control systems to optimize their usage and approach was used to enhance occupancy prediction accuracy minimize building energy consumption [1]. In the past years, when the amounts of historical training data are limited. The occupancy prediction modelling in buildings has been studied proposed approach and models are applied to a case study of by many researchers using different types of data such as air three office rooms in an educational building. The data sets used temperature, sound, door status, relative humidity, camera, in this work are actual data collected from the Urban Sciences motion and light.
    [Show full text]
  • Stylegans and Transfer Learning for Generating Synthetic Images in Industrial Applications
    S S symmetry Article StyleGANs and Transfer Learning for Generating Synthetic Images in Industrial Applications Harold Achicanoy 1,2,*,† , Deisy Chaves 2,3,*,† and Maria Trujillo 2,† 1 Alliance of Bioversity International and CIAT, Km 17 Recta Cali-Palmira, Palmira 763537, Colombia 2 School of Computer and Systems Engineering, Universidad del Valle, Cali 760001, Colombia; [email protected] 3 Department of Electrical, Systems and Automation, Universidad de León, 24007 León, Spain * Correspondence: [email protected] (H.A.); [email protected] (D.C.) † These authors contributed equally to this work. Abstract: Deep learning applications on computer vision involve the use of large-volume and repre- sentative data to obtain state-of-the-art results due to the massive number of parameters to optimise in deep models. However, data are limited with asymmetric distributions in industrial applications due to rare cases, legal restrictions, and high image-acquisition costs. Data augmentation based on deep learning generative adversarial networks, such as StyleGAN, has arisen as a way to create train- ing data with symmetric distributions that may improve the generalisation capability of built models. StyleGAN generates highly realistic images in a variety of domains as a data augmentation strategy but requires a large amount of data to build image generators. Thus, transfer learning in conjunction with generative models are used to build models with small datasets. However, there are no reports on the impact of pre-trained generative models, using transfer learning. In this paper, we evaluate a StyleGAN generative model with transfer learning on different application domains—training Citation: Achicanoy, H.; Chaves, D.; with paintings, portraits, Pokémon, bedrooms, and cats—to generate target images with different Trujillo, M.
    [Show full text]
  • Deep Learning Occupancy Activity Detection Approach for Optimising Building Energy Loads
    International Conference on Applied Energy 2020 Dec. 1 - Dec. 10, 2020, Bangkok / Virtual Paper ID: 127 Deep learning occupancy activity detection approach for optimising building energy loads Paige Wenbin Tien*, Shuangyu Wei, John Kaiser Calautit, Jo Darkwa, Christopher Wood Department of Architecture and Built Environment, University of Nottingham, Nottingham, United Kingdom *e-mail: [email protected], [email protected] ABSTRACT Heating, ventilation and air- HVAC The main aim of this paper is to develop a vision- conditioning based deep learning method for real-time occupancy activity detection and recognition to help the operations 1. INTRODUCTION AND LITERATURE REVIEW of building energy systems. A faster region-based The built environment sector is responsible for up to convolutional neural network was developed, trained 35% of the global energy use and energy-related and deployed to an artificial intelligence (AI)-powered emissions [1]. Reducing the energy consumption of camera for the application of real-time occupancy buildings requires innovative methods. Solutions such as activity detection. Initial experimental tests were occupancy-based controls can achieve significant energy performed within an office space of a selected case study savings by eliminating unnecessary energy usage. building. The detection provided correct detections for A significant element affecting the usage of these the majority of the time (97.32%). Average detection energy consumers is the behaviour of the occupants [2]. accuracy of 92.20% was achieved for all activities. For instance, rooms in offices or lecture theatres are not Building energy simulation of the case study building was fully utilised or occupied during the day.
    [Show full text]
  • Use Transfer Learning for Efficient Deep Learning Training on Intel® Xeon® Processors
    WHITE PAPER Artificial Intelligence Transfer Learning Intel AI® Builders Use Transfer Learning For Efficient Deep Learning Training On Intel® Xeon® Processors Authors Introduction Beenish Zia This is an educational white paper on transfer learning, showcasing how existing Intel Corporation deep learning models can be easily and flexibly customized to solve new problems. One of the biggest challenges with deep learning is the large number Ramesh Illikkal of labeled data points that are required to train the deep learning models to Intel Corporation sufficient accuracy. For example, the ImageNet*20 database for image recognition Bob Rogers consists of over 14 million hand-labeled images. While the number of possible Intel Corporation applications of deep learning systems in vision tasks, text processing, speech-to- text translation and many other domains is enormous, very few potential users of deep learning systems have sufficient training data to create models from scratch. A common concern among teams considering the use of deep learning to solve business problems is the need for training data: "Doesn't deep learning need Table of Contents millions of samples and months of training to get good results?" One powerful Introduction . 1 solution is transfer learning, in which part of an existing deep learning model is re-optimized on a small data set to solve a related, but new, problem. In fact, What Is Transfer Learning . .. 2 one of the great attractions of transfer learning is that, unlike most traditional Benefits Using Transfer Learning . 3 approaches to machine learning, we can take models trained on one (perhaps very large) dataset and modify them quickly and easily to work well on a new problem Applications Transfer Learning .
    [Show full text]
  • Deep Learning for EMG-Based Human-Machine Interaction: a Review Dezhen Xiong, Daohui Zhang, Member, IEEE, Xingang Zhao, Member, IEEE, and Yiwen Zhao
    512 IEEE/CAA JOURNAL OF AUTOMATICA SINICA, VOL. 8, NO. 3, MARCH 2021 Deep Learning for EMG-based Human-Machine Interaction: A Review Dezhen Xiong, Daohui Zhang, Member, IEEE, Xingang Zhao, Member, IEEE, and Yiwen Zhao Abstract—Electromyography (EMG) has already been broadly intramuscular EMG (iEMG), according to the electrodes’ used in human-machine interaction (HMI) applications. location. The former is collected from the surface of human Determining how to decode the information inside EMG signals robustly and accurately is a key problem for which we urgently skin, while the latter is collected from needle electrodes need a solution. Recently, many EMG pattern recognition tasks planted inside the human muscle. sEMG has been widely used have been addressed using deep learning methods. In this paper, for hand gesture classification [2], [3], silent speech recogni- we analyze recent papers and present a literature review tion [4], [5], stroke rehabilitation [6], [7], robot control [8], describing the role that deep learning plays in EMG-based HMI. [9], and other applications, mainly because it is cheap and An overview of typical network structures and processing schemes will be provided. Recent progress in typical tasks such as easy to collect and it provides a method for more natural movement classification, joint angle prediction, and force/torque human-machine collaboration. estimation will be introduced. New issues, including multimodal Many approaches, such as video, inertial measurement units sensing, inter-subject/inter-session, and robustness toward (IMU), and EMG, can be used to decode the movement disturbances will be discussed. We attempt to provide a comprehensive analysis of current research by discussing the intention of humans.
    [Show full text]
  • Transfer Learning and Intelligence: an Argument and Approach
    In The First Conference on Artificial General Intelligence (AGI-08), Memphis, Tennessee, March 2008. Transfer Learning and Intelligence: an Argument and Approach Matthew E. TAYLOR, Gregory KUHLMANN, and Peter STONE Department of Computer Sciences The University of Texas at Austin {mtaylor, kuhlmann, pstone}@cs.utexas.edu Abstract. In order to claim fully general intelligence in an autonomous agent, the ability to learn is one of the most central capabilities. Classical machine learn- ing techniques have had many significant empirical successes, but large real-world problems that are of interest to generally intelligent agents require learning much faster (with much less training experience) than is currently possible. This paper presents transfer learning, where knowledge from a learned task can be used to significantly speed up learning in a novel task, as the key to achieving the learning capabilities necessary for general intelligence. In addition to motivating the need for transfer learning in an intelligent agent, we introduce a novel method for select- ing types of tasks to be used for transfer and empirically demonstrate that such a selection can lead to significant increases in training speed in a two-player game. Keywords. Transfer Learning, Game Tree Search, Reinforcement Learning 1. Introduction A generally intelligent agent deployed in any non-trivial environment must be able to learn: no designer can anticipate every possible encountered situation. Specifically, in- telligent agents need to learn how to act, which is the purview of Reinforcement learn- ing [8,15]. Reinforcement learning (RL) problems are defined as those in which learn- ing agents sequentially execute actions with the goal of maximizing a reward signal, which may be time-delayed.
    [Show full text]