Offline Reinforcement Learning from Datasets with Structured Non-Stationarity.

Johannes Ackermann Takayuki Osa Masashi Sugiyama

Published in: CoRR (2024)

Keyphrases

reinforcement learning
function approximation
non stationary
uci machine learning repository
real world
learning process
structured data
decision trees
state space
data mining
machine learning
knowledge base
optimal policy
transfer learning
synthetic and real datasets
transition model