1
0
mirror of https://github.com/gryf/coach.git synced 2025-12-18 11:40:18 +01:00
Files
coach/benchmarks/dueling_ddqn

Dueling DDQN

Each experiment uses 3 seeds and is trained for 10k environment steps. The parameters used for Dueling DDQN are the same parameters as described in the original paper.

Pong Dueling DDQN - single worker

coach -p Atari_Dueling_DDQN -lvl pong
Pong Dueling DDQN

Breakout Dueling DDQN - single worker

coach -p Atari_Dueling_DDQN -lvl breakout
Breakout Dueling DDQN

Space Invaders Dueling DDQN - single worker

coach -p Atari_Dueling_DDQN -lvl space_invaders
Space Invaders Dueling DDQN