Deep Reinforcement Learning

The repository contains solutions for the projects in the Udacity course https://eu.udacity.com/course/deep-reinforcement-learning-nanodegree--nd893

Navigation Project

• Value Based Reinforcement Learning: Neural Network used to learn an approximation of the Q function (Deep-Q Learning).

• Details: https://github.com/antoniopenta/deep_reinforcement_learning/tree/master/drl_navigation

• Demo: https://www.youtube.com/watch?v=PwsPZzcfGG4

Continuos Control Project

• Policy Gradient Reinforcement Learning: Neural Network used to learn an approximation of the Policy function for continous actions (Deep Deterministic Policy Gradient-DDPG).

• Details: https://github.com/antoniopenta/deep_reinforcement_learning/tree/master/drl_continous_control

• Demo: https://www.youtube.com/watch?v=ryAWEyv8bm4

Multi Agent Reinforcement Learning

• Actor Critic based solution for collaboration between two agents.

• Details: https://github.com/antoniopenta/deep_reinforcement_learning/tree/master/drl_multiagent_rl

• Demo: https://www.youtube.com/watch?v=jtyQqDcZgRs

Donations

• If you found this code useful, please condider to make a donation to AVSI
  • https://www.avsi.org/en/country/
  • http://www.avsi-usa.org/donate.html
  • https://www.avsi.org/it/donation/

Many thanks!