Exploration vs Exploitation vs Safety: Risk-averse Multi-Armed Bandits.

Nicolas Galichet Michèle Sebag Olivier Teytaud

Published in: CoRR (2014)

Keyphrases

risk averse
multi armed bandits
risk neutral
risk aversion
utility function
stochastic programming
decision makers
bandit problems
portfolio management
expected utility
bayesian networks
multistage
decision making
dynamic programming
decision problems