Black-box policy search with probabilistic programs
In this work we show how to represent policies as programs: that is, as stochastic simulators with tunable parameters. To learn the parameters of such policies we develop connections between black box variational inference and existing policy search approaches. We then explain how such learning can...
Prif Awduron: | Van De Meent, J, Paige, B, Tolpin, D, Wood, F |
---|---|
Fformat: | Conference item |
Cyhoeddwyd: |
Journal of Machine Learning Research
2016
|
Eitemau Tebyg
-
Output-sensitive Adaptive Metropolis-Hastings for probabilistic programs
gan: Tolpin, D, et al.
Cyhoeddwyd: (2015) -
Maximum a posteriori estimation by search in probabilistic programs
gan: Tolpin, D, et al.
Cyhoeddwyd: (2015) -
Bayesian Optimization for Probabilistic Programs
gan: Rainforth, T, et al.
Cyhoeddwyd: (2016) -
Particle Gibbs with Ancestor Sampling for Probabilistic Programs
gan: van de Meent, J, et al.
Cyhoeddwyd: (2015) -
Particle Gibbs with ancestor sampling for probabilistic programs
gan: Meent, J, et al.
Cyhoeddwyd: (2015)