Exploring the Pareto front of multi-objective COVID-19 mitigation policies using reinforcement learning.
Mathieu ReymondConor F. HayesLander WillemRoxana RadulescuSteven AbramsDiederik M. RoijersEnda HowleyPatrick MannionNiel HensAnn NowéPieter LibinPublished in: Expert Syst. Appl. (2024)
Keyphrases
- multi objective
- reinforcement learning
- optimal policy
- multi objective optimization
- policy search
- evolutionary algorithm
- optimization algorithm
- multiobjective optimization
- markov decision process
- control policies
- nsga ii
- markov decision problems
- reward function
- partially observable markov decision processes
- reinforcement learning agents
- particle swarm optimization
- state space
- fitted q iteration
- genetic algorithm
- hierarchical reinforcement learning
- policy gradient methods
- multiple objectives
- pareto fronts
- dynamic programming
- machine learning
- learning algorithm
- objective function
- function approximation
- markov decision processes
- reinforcement learning algorithms
- control policy
- pareto optimal
- engineering design problems
- temporal difference
- model free
- multi objective optimization problems
- optimum design
- multi agent
- pareto optimal solutions
- conflicting objectives
- continuous state
- risk management
- partially observable
- optimal control
- multi objective evolutionary algorithms
- fitness function
- optimization problems
- multi agent reinforcement learning
- bi objective