Group Fairness in Reinforcement Learning via Multi-Objective Rewards.
Jack BlandinIan A. KashPublished in: Trans. Mach. Learn. Res. (2024)
Keyphrases
- reinforcement learning
- multi objective
- evolutionary algorithm
- markov decision processes
- multi objective optimization
- optimization algorithm
- function approximation
- genetic algorithm
- multiple objectives
- objective function
- state space
- particle swarm optimization
- multiobjective optimization
- reinforcement learning algorithms
- reward function
- model free
- multi agent
- conflicting objectives
- pareto optimal
- multi objective evolutionary
- optimal policy
- learning algorithm
- optimal control
- partially observable
- trade off
- reward shaping
- learning classifier systems
- game theory
- temporal difference
- transfer learning
- learning process
- bi objective
- control policy
- markov decision problems
- hidden state