Multi-Agent Meta-Reinforcement Learning: Sharper Convergence Rates with Task Similarity.
Weichao MaoHaoran QiuChen WangHubertus FrankeZbigniew KalbarczykRavishankar K. IyerTamer BasarPublished in: NeurIPS (2023)
Keyphrases
- convergence rate
- reinforcement learning
- multi agent
- step size
- learning rate
- convergence speed
- similarity measure
- function approximation
- reinforcement learning algorithms
- mutation operator
- state space
- multi agent environments
- primal dual
- temporal difference
- conjugate gradient
- numerical stability
- model free
- multi agent systems
- global convergence
- reinforcement learning agents
- learning algorithm
- gaussian kernels
- low memory requirements
- markov decision processes
- policy iteration
- single agent
- multiple agents
- high dimensional
- feature vectors
- transfer learning