GP-select: accelerating EM using adaptive subspace preselection
From MaRDI portal
Abstract: We propose a nonparametric procedure to achieve fast inference in generative graphical models when the number of latent states is very large. The approach is based on iterative latent variable preselection, where we alternate between learning a 'selection function' to reveal the relevant latent variables, and use this to obtain a compact approximation of the posterior distribution for EM; this can make inference possible where the number of possible latent states is e.g. exponential in the number of latent variables, whereas an exact approach would be computationally unfeasible. We learn the selection function entirely from the observed data and current EM state via Gaussian process regression. This is by contrast with earlier approaches, where selection functions were manually-designed for each problem setting. We show that our approach performs as well as these bespoke selection functions on a wide variety of inference problems: in particular, for the challenging case of a hierarchical model for object localization with occlusion, we achieve results that match a customized state-of-the-art selection method, at a far lower computational cost.
Recommendations
- Expectation truncation and the benefits of preselection in training generative models
- Generic inference in latent Gaussian process models
- Graphical models, exponential families, and variational inference
- An introduction to variational methods for graphical models
- scientific article; zbMATH DE number 1222283
Cites work
- scientific article; zbMATH DE number 6378149 (Why is no real title available?)
- scientific article; zbMATH DE number 5968904 (Why is no real title available?)
- scientific article; zbMATH DE number 3567782 (Why is no real title available?)
- scientific article; zbMATH DE number 1222290 (Why is no real title available?)
- 10.1162/15324430260185619
- Bayesian optimization for likelihood-free inference of simulator-based statistical models
- Design and analysis of computer experiments. With comments and a rejoinder by the authors
- Divide and conquer kernel ridge regression: a distributed algorithm with minimax optimal rates
- Efficient occlusive components analysis
- Expectation truncation and the benefits of preselection in training generative models
- Gaussian processes for machine learning.
- Online learning for matrix factorization and sparse coding
Cited in
(4)- Expectation truncation and the benefits of preselection in training generative models
- Linear and nonlinear generative models for `zero-shot' image denoising in the limit of few photons
- Neural Simpletrons: Learning in the Limit of Few Labels with Directed Generative Networks
- scientific article; zbMATH DE number 7625174 (Why is no real title available?)
This page was built for publication: GP-select: accelerating EM using adaptive subspace preselection
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5380836)
