GP-Tree: A Gaussian Process Classifier for Few-Shot Incremental Learning

TL;DR

GP-Tree introduces a hierarchical Gaussian process model combined with deep kernel learning, enabling scalable and accurate multi-class classification in few-shot incremental learning scenarios.

Contribution

It presents a novel tree-based hierarchical GP model that scales efficiently with data and class size, improving incremental few-shot learning performance.

Findings

Outperforms existing GP training baselines.

Achieves higher accuracy on incremental few-shot benchmarks.

Scales well with data size and number of classes.

Abstract

Gaussian processes (GPs) are non-parametric, flexible, models that work well in many tasks. Combining GPs with deep learning methods via deep kernel learning (DKL) is especially compelling due to the strong representational power induced by the network. However, inference in GPs, whether with or without DKL, can be computationally challenging on large datasets. Here, we propose GP-Tree, a novel method for multi-class classification with Gaussian processes and DKL. We develop a tree-based hierarchical model in which each internal node of the tree fits a GP to the data using the P\'olya Gamma augmentation scheme. As a result, our method scales well with both the number of classes and data size. We demonstrate the effectiveness of our method against other Gaussian process training baselines, and we show how our general GP approach achieves improved accuracy on standard incremental few-shot learning benchmarks.