Evolutionary-Guided Synthesis of Verified Pareto-Optimal MDP Policies.
Simos GerasimouJavier CámaraRadu CalinescuNaif AlasmariFaisal AlhwikemXinwei FangPublished in: ASE (2021)
Keyphrases
- pareto optimal
- optimal policy
- markov decision process
- expected utility
- multi objective
- markov decision problems
- markov decision processes
- multi objective optimization
- reward function
- multi issue negotiation
- state space
- multiple objectives
- genetic algorithm
- reinforcement learning
- nash equilibrium
- decision problems
- pareto optimality
- nsga ii
- discounted reward
- pareto optimal set
- evolutionary computation
- dynamic programming
- optimal solution
- social welfare
- evolutionary algorithm
- multiobjective optimization
- average cost
- partially observable markov decision processes
- average reward
- cooperative
- finite state
- particle swarm optimization
- long run
- pareto optimal solutions
- multi agent
- sufficient conditions
- greedy algorithm
- objective function
- linear programming