Skip to content

Latest commit

 

History

History
8 lines (6 loc) · 305 Bytes

README.md

File metadata and controls

8 lines (6 loc) · 305 Bytes

rl-falling-blocks

Building a few RL implementations on my falling blocks game.

Imitation learning

Imitation learning algorithm based on my own game play of the game (used as a benchmark).

Policy gradient

Basic REINFORCE algorithm with average reward baseline for unbiased (high variance) agent.