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...
Главные авторы: | Van De Meent, J, Paige, B, Tolpin, D, Wood, F |
---|---|
Формат: | Conference item |
Опубликовано: |
Journal of Machine Learning Research
2016
|
Схожие документы
-
Output-sensitive Adaptive Metropolis-Hastings for probabilistic programs
по: Tolpin, D, и др.
Опубликовано: (2015) -
Maximum a posteriori estimation by search in probabilistic programs
по: Tolpin, D, и др.
Опубликовано: (2015) -
Bayesian Optimization for Probabilistic Programs
по: Rainforth, T, и др.
Опубликовано: (2016) -
Particle Gibbs with Ancestor Sampling for Probabilistic Programs
по: van de Meent, J, и др.
Опубликовано: (2015) -
Particle Gibbs with ancestor sampling for probabilistic programs
по: Meent, J, и др.
Опубликовано: (2015)