Teacher-apprentices RL (TARL): leveraging complex policy distribution through generative adversarial hypernetwork in reinforcement learning.
Shi Yuan TangAthirai A. IrissappaneFrans A. OliehoekJie ZhangPublished in: Auton. Agents Multi Agent Syst. (2023)
Keyphrases
- reinforcement learning
- optimal policy
- markov decision process
- rl algorithms
- action selection
- policy search
- markov decision processes
- action space
- function approximation
- learning process
- multi agent
- approximate dynamic programming
- actor critic
- state and action spaces
- model free
- function approximators
- reinforcement learning algorithms
- partially observable
- state space
- policy gradient
- reinforcement learning problems
- control policy
- partially observable domains
- learning algorithm
- policy evaluation
- temporal difference
- transfer learning
- temporal difference learning
- markov decision problems
- control policies
- partially observable environments
- complex domains
- control problems
- reward function
- average reward
- optimal control
- decision problems
- dynamic programming
- direct policy search
- model free reinforcement learning
- state dependent
- machine learning
- finite state
- complex environments
- agent learns
- generative model
- semi supervised