Modified Policy Iteration for Exponential Cost Risk Sensitive MDPs.
Yashaswini MurthyMehrdad MoharramiR. SrikantPublished in: L4DC (2023)
Keyphrases
- risk sensitive
- markov decision processes
- policy iteration
- average cost
- markov decision problems
- optimal policy
- finite state
- optimal control
- state space
- reinforcement learning
- factored mdps
- model free
- finite horizon
- infinite horizon
- total cost
- dynamic programming
- markov decision process
- average reward
- linear programming
- policy evaluation
- action space
- partially observable
- planning under uncertainty
- reward function
- reinforcement learning algorithms
- finite number
- long run
- initial state
- control policy
- decision processes
- linear program
- fixed point
- dynamical systems
- multi agent