Multi-Agent Actor-Critic Multitask Reinforcement Learning based on GTD(1) with Consensus.
Milos S. StankovicMarko BekoNemanja IlicSrdjan S. StankovicPublished in: CDC (2022)
Keyphrases
- actor critic
- reinforcement learning
- multi task
- multi agent
- transfer learning
- learning problems
- multitask learning
- learning tasks
- temporal difference
- reinforcement learning algorithms
- function approximation
- policy gradient
- state space
- approximate dynamic programming
- optimal policy
- multi class
- single agent
- model free
- policy iteration
- optimal control
- multi agent systems
- supervised learning
- feature selection
- learning algorithm
- machine learning
- rl algorithms
- neuro fuzzy
- partially observable markov decision processes
- reinforcement learning methods
- unsupervised learning
- pairwise
- average reward
- temporal difference learning
- semi supervised learning
- semi supervised
- learning process
- policy gradient methods