Bias-Awareness for Zero-Shot Learning the Seen and Unseen

TL;DR

This paper introduces a bias-aware approach for generalized zero-shot learning that reduces bias towards seen classes by mapping inputs to a semantic space and regularizing class probabilities, improving performance across benchmarks.

Contribution

The paper proposes a novel bias-aware learner with temperature scaling and entropy regularization for generalized zero-shot learning, applicable to various semantic information types.

Findings

Improved accuracy on four zero-shot learning benchmarks.

Effective bias mitigation towards unseen classes.

Versatile approach usable with different semantic data.

Abstract

Generalized zero-shot learning recognizes inputs from both seen and unseen classes. Yet, existing methods tend to be biased towards the classes seen during training. In this paper, we strive to mitigate this bias. We propose a bias-aware learner to map inputs to a semantic embedding space for generalized zero-shot learning. During training, the model learns to regress to real-valued class prototypes in the embedding space with temperature scaling, while a margin-based bidirectional entropy term regularizes seen and unseen probabilities. Relying on a real-valued semantic embedding space provides a versatile approach, as the model can operate on different types of semantic information for both seen and unseen classes. Experiments are carried out on four benchmarks for generalized zero-shot learning and demonstrate the benefits of the proposed bias-aware classifier, both as a stand-alone method or in combination with generated features.