Unsupervised Feature Learning with Emergent Data-Driven Prototypicality

TL;DR

This paper introduces an unsupervised method for feature learning in hyperbolic space that encodes image prototypicality through spatial positioning, improving instance selection and model robustness.

Contribution

It proposes a novel hyperbolic space-based unsupervised feature learning algorithm using sphere packing, capturing prototypicality directly from data without labels.

Findings

Images closer to the origin are more prototypical.

Hyperbolic space captures prototypicality more effectively than Euclidean.

Method improves sample efficiency and model robustness.

Abstract

Given an image set without any labels, our goal is to train a model that maps each image to a point in a feature space such that, not only proximity indicates visual similarity, but where it is located directly encodes how prototypical the image is according to the dataset. Our key insight is to perform unsupervised feature learning in hyperbolic instead of Euclidean space, where the distance between points still reflect image similarity, and yet we gain additional capacity for representing prototypicality with the location of the point: The closer it is to the origin, the more prototypical it is. The latter property is simply emergent from optimizing the usual metric learning objective: The image similar to many training instances is best placed at the center of corresponding points in Euclidean space, but closer to the origin in hyperbolic space. We propose an unsupervised feature learning algorithm in Hyperbolic space with sphere pACKing. HACK first generates uniformly packed particles in the Poincar\'e ball of hyperbolic space and then assigns each image uniquely to each particle. Images after congealing are regarded more typical of the dataset it belongs to. With our feature mapper simply trained to spread out training instances in hyperbolic space, we observe that images move closer to the origin with congealing, validating our idea of unsupervised prototypicality discovery. We demonstrate that our data-driven prototypicality provides an easy and superior unsupervised instance selection to reduce sample complexity, increase model generalization with atypical instances and robustness with typical ones.