Identifying Policy Gradient Subspaces.

Jan Schneider Pierre Schumacher Simon Guist Le Chen Daniel F. B. Haeufle Bernhard Schölkopf Dieter Büchler

Published in: ICLR (2024)

Keyphrases

policy gradient
parametric optimization
actor critic
function approximation
gradient method
high dimensional data
high dimensional
principal component analysis
approximation methods
model free reinforcement learning
machine learning
feature space
single agent
partially observable markov decision processes