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...
Những tác giả chính: | Van De Meent, J, Paige, B, Tolpin, D, Wood, F |
---|---|
Định dạng: | Conference item |
Được phát hành: |
Journal of Machine Learning Research
2016
|
Những quyển sách tương tự
-
Output-sensitive Adaptive Metropolis-Hastings for probabilistic programs
Bằng: Tolpin, D, et al.
Được phát hành: (2015) -
Maximum a posteriori estimation by search in probabilistic programs
Bằng: Tolpin, D, et al.
Được phát hành: (2015) -
Bayesian Optimization for Probabilistic Programs
Bằng: Rainforth, T, et al.
Được phát hành: (2016) -
Particle Gibbs with Ancestor Sampling for Probabilistic Programs
Bằng: van de Meent, J, et al.
Được phát hành: (2015) -
Particle Gibbs with ancestor sampling for probabilistic programs
Bằng: Meent, J, et al.
Được phát hành: (2015)