Value Function Approximations via Kernel Embeddings for No-Regret Reinforcement Learning.
Sayak Ray ChowdhuryRafael OliveiraPublished in: ACML (2022)
Keyphrases
- reinforcement learning
- function approximators
- kernel methods
- function approximation
- kernel function
- markov decision processes
- support vector
- lower bound
- state space
- online learning
- machine learning
- reproducing kernel hilbert space
- reinforcement learning algorithms
- loss function
- approximation methods
- control policy
- hilbert space
- kernel matrix
- game theory
- low dimensional
- worst case
- multi agent
- learning algorithm