Bayesian Few-Shot Classification with One-vs-Each Pólya-Gamma Augmented Gaussian Processes

Citations Over Time

Abstract

Few-shot classification (FSC), the task of adapting a classifier to unseen classes given a small labeled dataset, is an important step on the path toward human-like machine learning. Bayesian methods are well-suited to tackling the fundamental issue of overfitting in the few-shot scenario because they allow practitioners to specify prior beliefs and update those beliefs in light of observed data. Contemporary approaches to Bayesian few-shot classification maintain a posterior distribution over model parameters, which is slow and requires storage that scales with model size. Instead, we propose a Gaussian process classifier based on a novel combination of Pólya-Gamma augmentation and the one-vs-each softmax approximation that allows us to efficiently marginalize over functions rather than model parameters. We demonstrate improved accuracy and uncertainty quantification on both standard few-shot classification benchmarks and few-shot domain transfer tasks.

Related Papers

• → Skip Residual Pairwise Networks With Learnable Comparative Functions for Few-Shot Learning(2019)1 cited
• → Transductive Information Maximization For Few-Shot Learning(2020)49 cited
• → Model Selection for Generalized Zero-shot Learning(2018)5 cited
• → GP-Tree: A Gaussian Process Classifier for Few-Shot Incremental Learning(2021)5 cited