Neighbour-Driven Gaussian Process Variational Autoencoders for Scalable Structured Latent Modelling
This work addresses scalability issues in structured latent modelling for researchers and practitioners in machine learning, representing an incremental improvement over existing GPVAE methods.
The paper tackles the computational challenge of performing exact Gaussian Process inference in large-scale GP Variational Autoencoders by proposing a neighbour-driven approximation strategy that exploits local adjacencies in the latent space. The result is improved predictive performance and computational efficiency, as demonstrated through experiments on tasks like representation learning and data imputation.
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.