# Scalable Training of Inference Networks for Gaussian-Process Models

arXiv: https://arxiv.org/abs/1905.10969 · HTML: https://mlsanity.com/paper/1905.10969/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 May 2019.

## Built on by (most cited first, 8 of 8)

- [The Gaussian Neural Process](https://mlsanity.com/paper/2101.03606/) (2021)
- [The Neural Process Family: Survey, Applications and Perspectives](https://mlsanity.com/paper/2209.00517/) (2022)
- [Fully Bayesian Autoencoders with Latent Sparse Gaussian Processes](https://mlsanity.com/paper/2302.04534/) (2023)
- [Input Dependent Sparse Gaussian Processes](https://mlsanity.com/paper/2107.07281/) (2021)
- [Deep Gaussian Processes for Biogeophysical Parameter Retrieval and Model Inversion](https://mlsanity.com/paper/2104.10638/) (2021)
- [Conditional deep generative models as surrogates for spatial field solution reconstruction with quantified uncertainty in Structural Health Monitoring applications](https://mlsanity.com/paper/2302.08329/) (2023)
- [Understanding Uncertainty in Bayesian Deep Learning](https://mlsanity.com/paper/2106.13055/) (2021)
- [Function-Space Regularization for Deep Bayesian Classification](https://mlsanity.com/paper/2307.06055/) (2023)

## Abstract (arXiv)

Inference in Gaussian process (GP) models is computationally challenging for large data, and often difficult to approximate with a small number of inducing points. We explore an alternative approximation that employs stochastic inference networks for a flexible inference. Unfortunately, for such networks, minibatch training is difficult to be able to learn meaningful correlations over function outputs for a large dataset. We propose an algorithm that enables such training by tracking a stochastic, functional mirror-descent algorithm. At each iteration, this only requires considering a finite number of input locations, resulting in a scalable and easy-to-implement algorithm. Empirical results show comparable and, sometimes, superior performance to existing sparse variational GP methods.