Bayesian Few-Shot Classification with One-vs-Each Pólya-Gamma Augmented Gaussian Processes
This work addresses the challenge of adapting classifiers to unseen classes with limited data, which is incremental as it builds on existing Bayesian methods by improving efficiency and performance.
The paper tackles the problem of overfitting in few-shot classification by proposing a Gaussian process classifier that uses Pólya-Gamma augmentation and one-vs-each softmax approximation to efficiently marginalize over functions, resulting in improved accuracy and uncertainty quantification on standard benchmarks and domain transfer tasks.
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.