GitPlanet

Cheap and reliable Node.js hosting starts at $3/month, and $1/month static HTML hosting

Created with love in Canada, visit hostnodejs.com today

Feel like to post an Ad? Learn Details
All Projects → SuReLI → Deep-RL-agents

SuReLI / Deep-RL-agents

Licence: MIT license
No description or website provided.

Programming Languages

python
139335 projects - #7 most used programming language
Jupyter Notebook
11667 projects

Labels

reinforcement-learningdeep-reinforcement-learning

Projects that are alternatives of or similar to Deep-RL-agents

rtrl
PyTorch implementation of our paper Real-Time Reinforcement Learning (NeurIPS 2019)
Stars: ✭ 57 (+111.11%)
Mutual labels:  deep-reinforcement-learning
pytorch-hdqn
Hierarchical-DQN in pytorch (not actively maintained)
Stars: ✭ 36 (+33.33%)
Mutual labels:  deep-reinforcement-learning
FLEXS
Fitness landscape exploration sandbox for biological sequence design.
Stars: ✭ 92 (+240.74%)
Mutual labels:  deep-reinforcement-learning
semantic-guidance
Code for our CVPR-2021 paper on Combining Semantic Guidance and Deep Reinforcement Learning For Generating Human Level Paintings.
Stars: ✭ 19 (-29.63%)
Mutual labels:  deep-reinforcement-learning
Smart-Traffic-Signals-in-India-using-Deep-Reinforcement-Learning-and-Advanced-Computer-Vision
We have used Deep Reinforcement Learning and Advanced Computer Vision techniques to for the creation of Smart Traffic Signals for Indian Roads. We have created the scripts for using SUMO as our environment for deploying all our RL models.
Stars: ✭ 131 (+385.19%)
Mutual labels:  deep-reinforcement-learning
deep reinforcement learning gallery
Deep reinforcement learning with tensorflow2
Stars: ✭ 35 (+29.63%)
Mutual labels:  deep-reinforcement-learning
pokeai
Develop ultimate AI Pokémon trainer
Stars: ✭ 18 (-33.33%)
Mutual labels:  deep-reinforcement-learning
MaRLEnE
Machine- and Reinforcement Learning ExtensioN for (game) Engines
Stars: ✭ 47 (+74.07%)
Mutual labels:  deep-reinforcement-learning
Reinforcement Learning Course
Curso de Aprendizaje por Refuerzo, de 0 a 100 con notebooks y slides muy sencillas para entenderlo todo perfectamente.
Stars: ✭ 18 (-33.33%)
Mutual labels:  deep-reinforcement-learning
rl-medical
Communicative Multiagent Deep Reinforcement Learning for Anatomical Landmark Detection using PyTorch.
Stars: ✭ 36 (+33.33%)
Mutual labels:  deep-reinforcement-learning
Underflow
With underflow, create trafic light clusters that interact together to regulate circulation
Stars: ✭ 12 (-55.56%)
Mutual labels:  deep-reinforcement-learning
AlphaNPI
Adapting the AlphaZero algorithm to remove the need of execution traces to train NPI.
Stars: ✭ 71 (+162.96%)
Mutual labels:  deep-reinforcement-learning
Deep-Quality-Value-Family
Official implementation of the paper "Approximating two value functions instead of one: towards characterizing a new family of Deep Reinforcement Learning Algorithms": https://arxiv.org/abs/1909.01779 To appear at the next NeurIPS2019 DRL-Workshop
Stars: ✭ 12 (-55.56%)
Mutual labels:  deep-reinforcement-learning
racing dreamer
Latent Imagination Facilitates Zero-Shot Transfer in Autonomous Racing
Stars: ✭ 31 (+14.81%)
Mutual labels:  deep-reinforcement-learning
playing-mario-with-deep-reinforcement-learning
An implementation of (Double/Dueling) Deep-Q Learning to play Super Mario Bros.
Stars: ✭ 55 (+103.7%)
Mutual labels:  deep-reinforcement-learning
AutoPentest-DRL
AutoPentest-DRL: Automated Penetration Testing Using Deep Reinforcement Learning
Stars: ✭ 196 (+625.93%)
Mutual labels:  deep-reinforcement-learning
Deep-Reinforcement-Learning-CS285-Pytorch
Solutions of assignments of Deep Reinforcement Learning course presented by the University of California, Berkeley (CS285) in Pytorch framework
Stars: ✭ 104 (+285.19%)
Mutual labels:  deep-reinforcement-learning
Deep-Q-Networks
Implementation of Deep/Double Deep/Dueling Deep Q networks for playing Atari games using Keras and OpenAI gym
Stars: ✭ 38 (+40.74%)
Mutual labels:  deep-reinforcement-learning
deep-rl-quadcopter
Implementation of Deep Deterministic Policy Gradients (DDPG) to teach a Quadcopter How to Fly!
Stars: ✭ 17 (-37.04%)
Mutual labels:  deep-reinforcement-learning
robustnav
Evaluating pre-trained navigation agents under corruptions
Stars: ✭ 18 (-33.33%)
Mutual labels:  deep-reinforcement-learning
View All Similar Projects ➔

Deep-RL-agents

This repository contains the work I did during my Reinforcement Learning internship from September 2017 to February 2018.

During these 6 months, I reimplemented the main deep-RL algorithms that have been developped since 2013, using only Tensorflow and numpy. This repository contains implementations of :

  • A3C : the 2016 algorithm that uses asynchronous gradient descent for optimization on multi-CPU instead of a single GPU
  • C51 : the 2017 algorithm that explores the idea of predicting not only the value of a state, but instead the value distribution
  • DDPG : the 2015 algorithm that tackles the problem of continuous control using an actor-critic architecture
  • Rainbow : the 2017 algorithm that combines six classical extensions to DQN
  • D4PG : the 2018 algorithm that applies the distributional approach to a DDPG with an asynchronous architecture

The general architecture of these algorithm is always the same :

  • the main.py file initialize the agent and run it
  • the Model.py file implements the Neural Network (actor-critic or not, with convolution or not)
  • the QNetwork.py file instantiates a Network and build the tensorflow operations to perform the gradient descent to train it
  • the Agent.py file implements the agent class that interacts with the environment in order to get experiences
  • the settings.py file is used to change the hyperparameters of the algorithm and the network

Others directories include :

  • utils : a set of classes and functions used in other algorithms
  • BlogFiles : a jupyter notebook that tries to explain the idea behind A3C, DDPG and Rainbow
  • Environment Test : copies from the main algorithms set up to run in specific environments
  • GIF : a set of GIF saved after having trained different agents on many environments
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].