Natural Gradients in Practice: Non-Conjugate Variational Inference in Gaussian Process Models
نویسندگان
چکیده
The natural gradient method has been used effectively in conjugate Gaussian process models, but the non-conjugate case has been largely unexplored. We examine how natural gradients can be used in non-conjugate stochastic settings, together with hyperparameter learning. We conclude that the natural gradient can significantly improve performance in terms of wall-clock time. For illconditioned posteriors the benefit of the natural gradient method is especially pronounced, and we demonstrate a practical setting where ordinary gradients are unusable. We show how natural gradients can be computed efficiently and automatically in any parameterization, using automatic differentiation. Our code is integrated into the GPflow package.
منابع مشابه
Kullback-Leibler Proximal Variational Inference
We propose a new variational inference method based on a proximal framework that uses the Kullback-Leibler (KL) divergence as the proximal term. We make two contributions towards exploiting the geometry and structure of the variational bound. Firstly, we propose a KL proximal-point algorithm and show its equivalence to variational inference with natural gradients (e.g. stochastic variational in...
متن کاملFast Dual Variational Inference for Non-Conjugate Latent Gaussian Models
Latent Gaussian models (LGMs) are widely used in statistics and machine learning. Bayesian inference in non-conjugate LGMs is difficult due to intractable integrals involving the Gaussian prior and non-conjugate likelihoods. Algorithms based on variational Gaussian (VG) approximations are widely employed since they strike a favorable balance between accuracy, generality, speed, and ease of use....
متن کاملConjugate-Computation Variational Inference: Converting Variational Inference in Non-Conjugate Models to Inferences in Conjugate Models
Variational inference is computationally challenging in models that contain both conjugate and non-conjugate terms. Methods specifically designed for conjugate models, even though computationally efficient, find it difficult to deal with non-conjugate terms. On the other hand, stochastic-gradient methods can handle the nonconjugate terms but they usually ignore the conjugate structure of the mo...
متن کاملGray-box Inference for Structured Gaussian Process Models
We develop an automated variational inference method for Bayesian structured prediction problems with Gaussian process (gp) priors and linear-chain likelihoods. Our approach does not need to know the details of the structured likelihood model and can scale up to a large number of observations. Furthermore, we show that the required expected likelihood term and its gradients in the variational o...
متن کاملVariational Inference on Deep Exponential Family by using Variational Inferences on Conjugate Models
In this paper, we propose a new variational inference method for deep exponentialfamily (DEF) models. Our method converts non-conjugate factors in a DEF model to easy-to-compute conjugate exponential-family messages. This enables local and modular updates similar to variational message passing, as well as stochastic natural-gradient updates similar to stochastic variational inference. Such upda...
متن کامل