Bayesian Basis Function Approximation for Scalable Gaussian Process Priors in Deep Generative Models

Loading...
Thumbnail Image

Access rights

openAccess
CC BY-SA
publishedVersion

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Date

Major/Subject

Mcode

Degree programme

Language

en

Pages

24

Series

Proceedings of the 42nd International Conference on Machine Learning, pp. 2673-2696, Proceedings of Machine Learning Research ; Volume 267

Abstract

High-dimensional time-series datasets are common in domains such as healthcare and economics. Variational autoencoder (VAE) models, where latent variables are modeled with a Gaussian process (GP) prior, have become a prominent model class to analyze such correlated datasets. However, their applications are challenged by the inherent cubic time complexity that requires specific GP approximation techniques, as well as the general challenge of modeling both shared and individual-specific correlations across time. Though inducing points enhance GP prior VAE scalability, optimizing them remains challenging, especially since discrete covariates resist gradient‑based methods. In this work, we propose a scalable basis function approximation technique for GP prior VAEs that mitigates these challenges and results in linear time complexity, with a global parametrization that eliminates the need for amortized variational inference and the associated amortization gap, making it well-suited for conditional generation tasks where accuracy and efficiency are crucial. Empirical evaluations on synthetic and real-world benchmark datasets demonstrate that our approach not only improves scalability and interpretability but also drastically enhances predictive performance.

Description

Keywords

Other note

Citation

Balik, M Y, Sinelnikov, M, Ong, P & Lähdesmäki, H 2025, Bayesian Basis Function Approximation for Scalable Gaussian Process Priors in Deep Generative Models. in Proceedings of the 42nd International Conference on Machine Learning. Proceedings of Machine Learning Research, vol. 267, JMLR, pp. 2673-2696. < https://proceedings.mlr.press/v267/balik25a.html >