Mean-Semivariance Policy Optimization via Risk-Averse Reinforcement Learning (Extended Abstract).
Xiaoteng MaShuai MaLi XiaQianchuan ZhaoPublished in: IJCAI (2023)
Keyphrases
- extended abstract
- risk averse
- reinforcement learning
- stochastic programming
- optimal policy
- risk neutral
- markov decision process
- decision makers
- risk aversion
- function approximation
- function approximators
- markov decision processes
- multistage
- policy iteration
- partially observable
- control policies
- optimization problems
- state space
- utility function
- control policy
- expected utility
- asymptotically optimal
- markov decision problems
- long run
- model free
- infinite horizon
- optimization methods
- optimization method
- optimization algorithm
- linear programming
- dynamic programming
- objective function