Hindsight Expectation Maximization for Goal-conditioned Reinforcement Learning.

Yunhao Tang Alp Kucukelbir

Published in: AISTATS (2021)

Keyphrases

reinforcement learning
expectation maximization
em algorithm
image segmentation
generative model
optimal policy
markov decision processes
function approximation
machine learning
hidden markov models
dynamic programming
state space
mixture model
parameter estimation
action selection
temporal difference