-Review-Continuous-Control-With-Deep-Reinforcement-Learning | [Review and implementation code]

 by   170928 Python Version: Current License: No License

kandi X-RAY | -Review-Continuous-Control-With-Deep-Reinforcement-Learning Summary

kandi X-RAY | -Review-Continuous-Control-With-Deep-Reinforcement-Learning Summary

-Review-Continuous-Control-With-Deep-Reinforcement-Learning is a Python library typically used in Telecommunications, Media, Media, Entertainment applications. -Review-Continuous-Control-With-Deep-Reinforcement-Learning has no bugs, it has no vulnerabilities and it has low support. However -Review-Continuous-Control-With-Deep-Reinforcement-Learning build file is not available. You can download it from GitHub.

인공지능 분야의 중요한 목표 중 하나는 처리되지 않는 (unprocessed), 고 차원의 입력 데이터로부터 복잡한 작업을 해결해내는 것 입니다. 최근 "Deep Q Netwrok" (DQN) 알고리즘은 처리되지 않은 raw pixel의 입력으로부터 사람보다 뛰어난 Atari Game 을 플레이 하는 것을 보여주었습니다. 이때, 인공 신경망의 function approximation 능력은 강화학습의 "action-value function"을 대신하기위해서 사용됩니다. 그러나, DQN은 고차원의 observation space 문제를 해결하였지만, action space 측면에서는 저 차원, discrete handle의 경우만을 처리할 수 있었습니다. 문제는 대부분의 작업들이 continuous (real valued), high-dimensional 한 action space 를 가진다는 것입니다. DQN은 action-value fuction 을 최대화 하는 action을 찾는 것에만 집중하는 모델이기 때문에, 이러한 continous domain action에 적용되기 어렵습니다. Continuous action domains을 DQN에 적용하기 위해서 간단한 방법은 action space를 discretize 하는 것입니다. 그러나. 이 방법은 "Curse of dimensionality"라는 한계를 갖게 합니다. large action space 는 효과적은로 학습과정에서 탐색하기 어려울 뿐더러, 성공적인 학습이 어렵게 됩니다. 뿐만 아니라, action space의 discretization은 문제 해결에 중요한 정보인 action domain에 대한 구조적 정보를 사라지게 할 수 있습니다. 이 논문에서 제안하는 알고리즘은 "model-free, off-policy actor-critic" 알고리즘입니다. neural network의 function approximation 특징을 활용해서 고차원의 continuous action space에서의 정책 (Policy) 를 학습합니다. 제안하는 알고리즘은 deterministic policy gradient (DPG) 알고리즘을 기반으로 합니다. 그러나, 기존의 actor-critic method with neurla function approximators는 학습과정에서 안정적이지 못하다는 문제점이 있었습니다. 그래서, 이 논문은 이전 DQN에서 제시한 방법을 적용시켜서 성능을 향상 시켰습니다. (1) Replay buffer를 통해서 학습 데이터가 갖는 correlation을 줄입니다. (2) off-policy 를 통해서 네트워크를 학습합니다. (3) target Q network를 사용하여 Temporal Difference backups동안 target 값을 일정하게 유지합니다. (4) Batch normalization 기법을 활용합니다.
Support
    Quality
      Security
        License
          Reuse

            kandi-support Support

              -Review-Continuous-Control-With-Deep-Reinforcement-Learning has a low active ecosystem.
              It has 5 star(s) with 0 fork(s). There are no watchers for this library.
              OutlinedDot
              It had no major release in the last 6 months.
              There are 31 open issues and 0 have been closed. On average issues are closed in 781 days. There are no pull requests.
              It has a neutral sentiment in the developer community.
              The latest version of -Review-Continuous-Control-With-Deep-Reinforcement-Learning is current.

            kandi-Quality Quality

              -Review-Continuous-Control-With-Deep-Reinforcement-Learning has no bugs reported.

            kandi-Security Security

              -Review-Continuous-Control-With-Deep-Reinforcement-Learning has no vulnerabilities reported, and its dependent libraries have no vulnerabilities reported.

            kandi-License License

              -Review-Continuous-Control-With-Deep-Reinforcement-Learning does not have a standard license declared.
              Check the repository for any license declaration and review the terms closely.
              OutlinedDot
              Without a license, all rights are reserved, and you cannot use the library in your applications.

            kandi-Reuse Reuse

              -Review-Continuous-Control-With-Deep-Reinforcement-Learning releases are not available. You will need to build from source code and install.
              -Review-Continuous-Control-With-Deep-Reinforcement-Learning has no build file. You will be need to create the build yourself to build the component from source.

            Top functions reviewed by kandi - BETA

            kandi has reviewed -Review-Continuous-Control-With-Deep-Reinforcement-Learning and discovered the below as its top functions. This is intended to give you an instant insight into -Review-Continuous-Control-With-Deep-Reinforcement-Learning implemented functionality, and help decide if they suit your requirements.
            • Train a model
            • Sample a batch of data
            • Adds history to buffer
            • Subtract the discounted distributions from the prior distribution
            • Calculates the Vorstein upper air pressure based on the previous state
            • Return a random brownian response
            • Calculate the gradient of the action gradient
            • Return the number of records in the queue
            • Build the actor network
            • Create a bias variable
            • Create a weight variable
            • Build the network
            • Return the number of trainable variables
            Get all kandi verified functions for this library.

            -Review-Continuous-Control-With-Deep-Reinforcement-Learning Key Features

            No Key Features are available at this moment for -Review-Continuous-Control-With-Deep-Reinforcement-Learning.

            -Review-Continuous-Control-With-Deep-Reinforcement-Learning Examples and Code Snippets

            No Code Snippets are available at this moment for -Review-Continuous-Control-With-Deep-Reinforcement-Learning.

            Community Discussions

            No Community Discussions are available at this moment for -Review-Continuous-Control-With-Deep-Reinforcement-Learning.Refer to stack overflow page for discussions.

            Community Discussions, Code Snippets contain sources that include Stack Exchange Network

            Vulnerabilities

            No vulnerabilities reported

            Install -Review-Continuous-Control-With-Deep-Reinforcement-Learning

            You can download it from GitHub.
            You can use -Review-Continuous-Control-With-Deep-Reinforcement-Learning like any standard Python library. You will need to make sure that you have a development environment consisting of a Python distribution including header files, a compiler, pip, and git installed. Make sure that your pip, setuptools, and wheel are up to date. When using pip it is generally recommended to install packages in a virtual environment to avoid changes to the system.

            Support

            For any new features, suggestions and bugs create an issue on GitHub. If you have any questions check and ask questions on community page Stack Overflow .
            Find more information at:

            Find, review, and download reusable Libraries, Code Snippets, Cloud APIs from over 650 million Knowledge Items

            Find more libraries
            CLONE
          • HTTPS

            https://github.com/170928/-Review-Continuous-Control-With-Deep-Reinforcement-Learning.git

          • CLI

            gh repo clone 170928/-Review-Continuous-Control-With-Deep-Reinforcement-Learning

          • sshUrl

            git@github.com:170928/-Review-Continuous-Control-With-Deep-Reinforcement-Learning.git

          • Stay Updated

            Subscribe to our newsletter for trending solutions and developer bootcamps

            Agree to Sign up and Terms & Conditions

            Share this Page

            share link