Discrete Latent Space World Models for Reinforcement Learning

12 Oct 2020 Jan Robine Tobias Uelwer Stefan Harmeling

Sample efficiency remains a fundamental issue of reinforcement learning. Model-based algorithms try to make better use of data by simulating the environment with a model... (read more)

PDF Abstract
No code implementations yet. Submit your code now

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Atari Games Atari 2600 Bank Heist Discrete Latent Space World Model (VQ-VAE) Score 121.6 # 38
Atari Games Atari 2600 Breakout Discrete Latent Space World Model (VQ-VAE) Score 11.6 # 41
Atari Games Atari 2600 Crazy Climber Discrete Latent Space World Model (VQ-VAE) Score 59609.4 # 35
Atari Games Atari 2600 Freeway Discrete Latent Space World Model (VQ-VAE) Score 29 # 21
Atari Games Atari 2600 Pong Discrete Latent Space World Model (VQ-VAE) Score 20.2 # 7
Atari Games Atari 2600 Seaquest Discrete Latent Space World Model (VQ-VAE) Score 635 # 42

Methods used in the Paper