Overcoming model bias for robust offline deep reinforcement learning.

Phillip Swazinna Steffen Udluft Thomas A. Runkler

Published in: Eng. Appl. Artif. Intell. (2021)

Keyphrases

reinforcement learning
computational model
high level
probabilistic model
database
formal model
neural network
conceptual model
experimental data
statistical model
cost function
multi agent
theoretical analysis
mathematical model
learning process
expert systems
model free