Value Function Approximations via Kernel Embeddings for No-Regret Reinforcement Learning.

Sayak Ray Chowdhury, Rafael Oliveira

Browse the full ACML paper archive.