A likelihood-free Bayesian RL method trains deep generative environment models with prequential scoring rules, samples via SMC, and learns policies with expected Thompson sampling.
Asymptotics for power posterior mean estimation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Generalized Bayesian deep reinforcement learning
A likelihood-free Bayesian RL method trains deep generative environment models with prequential scoring rules, samples via SMC, and learns policies with expected Thompson sampling.