Risk Aversion in Markov Decision Processes via Near Optimal Chernoff Bounds.
Teodor Mihai MoldovanPieter AbbeelPublished in: NIPS (2012)
Keyphrases
- markov decision processes
- risk aversion
- utility function
- expected utility
- risk averse
- discounted reward
- optimal policy
- upper bound
- state space
- dynamic programming
- finite state
- reinforcement learning
- inventory level
- exchange rate
- lower bound
- policy iteration
- average reward
- infinite horizon
- partially observable
- action space
- worst case
- markov decision process
- decision theoretic
- decision makers
- risk sensitive
- non stationary
- long run
- decision theory
- average cost
- decision problems