Generalised entropy MDPs and Minimax Regret
Paper in proceeding, 2014
Bayesian methods suffer from the problem of how to specify prior beliefs.
One interesting idea is to consider worst-case priors. This requires solving
a stochastic zero-sum game. In this paper, we extend well-known results
from bandit theory in order to discover minimax-Bayes policies and discuss
when they are practical.
Author
Emmanouil Androulakis
Chalmers, Mathematical Sciences
University of Gothenburg
Christos Dimitrakakis
Chalmers, Computer Science and Engineering (Chalmers), Computing Science (Chalmers)
NIPS 2014, From bad models to good policies workshop.
Areas of Advance
Information and Communication Technology
Subject Categories
Probability Theory and Statistics