Risk-Averse Offline Reinforcement Learning.

Núria Armengol Urpí Sebastian Curi Andreas Krause

Published in: ICLR (2021)

Keyphrases

risk averse
reinforcement learning
risk neutral
risk aversion
decision makers
utility function
stochastic programming
function approximation
portfolio management
state space
expected utility
multistage
model free
markov decision processes
learning algorithm
dynamic programming
artificial intelligence