-
Notifications
You must be signed in to change notification settings - Fork 11
Home
Chua Cheow Huan edited this page Sep 2, 2019
·
6 revisions
Welcome to the reinforcement_learning wiki!
This repository contains codes that I have reproduced for various reinforcement learning algorithms. The codes were tested on Colab.
| Algorithms | Discrete | Continuous | Multithreaded | Multiprocessing | Tested on |
|---|---|---|---|---|---|
| DQN | ✔️ | CartPole-v0 | |||
| Double DQN (DDQN) | ✔️ | CartPole-v0 | |||
| Dueling DDQN | ✔️ | CartPole-v0 | |||
| Dueling DDQN + PER | ✔️ | CartPole-v0 | |||
| A3C (1) | ✔️ | ✔️ | ✔️ | ✔️(3) | CartPole-v0, Pendulum-v0 |
| DPPO (2) | ✔️ | ✔️(3) | Pendulum-v0 | ||
| RND + PPO | ✔️ | MountainCarContinuous-v0 |
(1): N-step returns used for critic's target.
(1): GAE used for computation of TD lambda return (for critic's target) & policy's advantage.
(3): Distributed Tensorflow & Python's multiprocessing package used.
Check out my blog for more information on my repositories.