/DQN-Algorithms

Implementation of DQN algorithm with extensions

Primary LanguagePython

SYDE522 Final Project

Initial results of Vanilla DQN trained with the pong environment.

Implemenentation of DQN training using PyTorch (uses a bit of starter code from https://github.com/berkeleydeeprlcourse/homework)

Exploration of various algorithms and implementations starting from what was presented in https://www.cs.toronto.edu/~vmnih/docs/dqn.pdf.

Additionally exploring the impact of learning with:

  1. Vanilla DQN (https://www.cs.toronto.edu/~vmnih/docs/dqn.pdf) (working)
  2. Multiple parallel agents (https://arxiv.org/pdf/1602.01783.pdf) (testing)
  3. Double DQN (https://arxiv.org/pdf/1509.06461.pdf) (testing)
  4. TBD.