Lexicographic Optimisation of Conditional Value at Risk and Expected Value for Risk-Averse Planning in MDPs.
Marc RigterPaul DuckworthBruno LacerdaNick HawesPublished in: CoRR (2021)
Keyphrases
- risk neutral
- risk averse
- utility function
- markov decision processes
- risk sensitive
- risk aversion
- markov decision problems
- decision makers
- planning problems
- expected utility
- stochastic programming
- reinforcement learning
- decision theoretic
- partially observable
- portfolio management
- state space
- initial state
- inventory level
- planning horizon
- combinatorial optimization
- dynamic programming
- genetic algorithm
- decision theory
- optimal control
- linear program
- optimal policy