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

TL;DR

This paper introduces a neighbour-driven approximation for Gaussian Process Variational Autoencoders that enables scalable inference by leveraging local adjacency, improving flexibility and efficiency over traditional methods.

Contribution

It proposes a novel neighbour-driven strategy for scalable GPVAE inference that preserves latent dependencies and allows flexible kernels without extensive inducing points.

Findings

Outperforms existing GPVAE variants in predictive accuracy

Achieves better computational efficiency in large-scale tasks

Enhances flexibility in kernel choices for latent modeling

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.