Learning the Compositional Spaces for Generalized Zero-shot Learning

TL;DR

This paper introduces a novel framework for Generalized Zero-shot Learning that decomposes the feature space into Source, Target, and Uncertain regions, enabling more accurate classification of seen and unseen classes.

Contribution

It proposes a new space decomposition method using bootstrapping and K-S test to learn and fine-tune decision boundaries between classes in G-ZSL.

Findings

Achieves state-of-the-art performance on multiple G-ZSL benchmarks.

Effectively separates instances into source, target, and uncertain spaces.

Improves classification accuracy for unseen classes.

Abstract

This paper studies the problem of Generalized Zero-shot Learning (G-ZSL), whose goal is to classify instances belonging to both seen and unseen classes at the test time. We propose a novel space decomposition method to solve G-ZSL. Some previous models with space decomposition operations only calibrate the confident prediction of source classes (W-SVM [46]) or take target-class instances as outliers [49]. In contrast, we propose to directly estimate and fine-tune the decision boundary between the source and the target classes. Specifically, we put forward a framework that enables to learn compositional spaces by splitting the instances into Source, Target, and Uncertain spaces and perform recognition in each space, where the uncertain space contains instances whose labels cannot be confidently predicted. We use two statistical tools, namely, bootstrapping and Kolmogorov-Smirnov (K-S) Test, to learn the compositional spaces for G-ZSL. We validate our method extensively on multiple G-ZSL benchmarks, on which it achieves state-of-the-art performances.