Neighbour-Driven Gaussian Process Variational Autoencoders for Scalable Structured Latent Modelling

Xinxing Shi, Xiaoyu Jiang, Mauricio A Álvarez
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:54934-54959, 2025.

Abstract

Gaussian Process (GP) Variational Autoencoders (VAEs) extend standard VAEs by replacing the fully factorised Gaussian prior with a GP prior, thereby capturing richer correlations among latent variables. However, performing exact GP inference in large-scale GPVAEs is computationally prohibitive, often forcing existing approaches to rely on restrictive kernel assumptions or large sets of inducing points. In this work, we propose a neighbour-driven approximation strategy that exploits local adjacencies in the latent space to achieve scalable GPVAE inference. By confining computations to the nearest neighbours of each data point, our method preserves essential latent dependencies, allowing more flexible kernel choices and mitigating the need for numerous inducing points. Through extensive experiments on tasks including representation learning, data imputation, and conditional generation, we demonstrate that our approach outperforms other GPVAE variants in both predictive performance and computational efficiency.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-shi25e, title = {Neighbour-Driven {G}aussian Process Variational Autoencoders for Scalable Structured Latent Modelling}, author = {Shi, Xinxing and Jiang, Xiaoyu and \'{A}lvarez, Mauricio A}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {54934--54959}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/shi25e/shi25e.pdf}, url = {https://proceedings.mlr.press/v267/shi25e.html}, abstract = {Gaussian Process (GP) Variational Autoencoders (VAEs) extend standard VAEs by replacing the fully factorised Gaussian prior with a GP prior, thereby capturing richer correlations among latent variables. However, performing exact GP inference in large-scale GPVAEs is computationally prohibitive, often forcing existing approaches to rely on restrictive kernel assumptions or large sets of inducing points. In this work, we propose a neighbour-driven approximation strategy that exploits local adjacencies in the latent space to achieve scalable GPVAE inference. By confining computations to the nearest neighbours of each data point, our method preserves essential latent dependencies, allowing more flexible kernel choices and mitigating the need for numerous inducing points. Through extensive experiments on tasks including representation learning, data imputation, and conditional generation, we demonstrate that our approach outperforms other GPVAE variants in both predictive performance and computational efficiency.} }
Endnote
%0 Conference Paper %T Neighbour-Driven Gaussian Process Variational Autoencoders for Scalable Structured Latent Modelling %A Xinxing Shi %A Xiaoyu Jiang %A Mauricio A Álvarez %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-shi25e %I PMLR %P 54934--54959 %U https://proceedings.mlr.press/v267/shi25e.html %V 267 %X Gaussian Process (GP) Variational Autoencoders (VAEs) extend standard VAEs by replacing the fully factorised Gaussian prior with a GP prior, thereby capturing richer correlations among latent variables. However, performing exact GP inference in large-scale GPVAEs is computationally prohibitive, often forcing existing approaches to rely on restrictive kernel assumptions or large sets of inducing points. In this work, we propose a neighbour-driven approximation strategy that exploits local adjacencies in the latent space to achieve scalable GPVAE inference. By confining computations to the nearest neighbours of each data point, our method preserves essential latent dependencies, allowing more flexible kernel choices and mitigating the need for numerous inducing points. Through extensive experiments on tasks including representation learning, data imputation, and conditional generation, we demonstrate that our approach outperforms other GPVAE variants in both predictive performance and computational efficiency.
APA
Shi, X., Jiang, X. & Álvarez, M.A.. (2025). Neighbour-Driven Gaussian Process Variational Autoencoders for Scalable Structured Latent Modelling. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:54934-54959 Available from https://proceedings.mlr.press/v267/shi25e.html.

Related Material