An Empirical Comparison of Off-policy Prediction Learning Algorithms in the Four Rooms Environment.

Sina Ghiassian Richard S. Sutton

Published in: CoRR (2021)

Keyphrases

learning algorithm
prediction accuracy
machine learning
mobile robot
machine learning algorithms

environmental conditions
data sets
dynamic environments
training data
upper bound

virtual world
multi agent
autonomous agents
learning tasks
learning problems

prediction error
indoor environments
genetic algorithm
efficient learning
general loss functions