Prototypical Model with Novel Information-theoretic Loss Function for Generalized Zero Shot Learning

TL;DR

This paper introduces a deterministic prototypical model with novel information-theoretic loss functions for generalized zero-shot learning, achieving state-of-the-art results and demonstrating compatibility with generative models.

Contribution

The paper proposes a new deterministic GZSL model using information-theoretic loss functions and probability vector representation, outperforming existing methods.

Findings

Achieves 21%-64% improvements over baseline models.

State-of-the-art results on GZSL benchmarks.

Compatible with generative models like f-CLSWGAN.

Abstract

Generalized zero shot learning (GZSL) is still a technical challenge of deep learning as it has to recognize both source and target classes without data from target classes. To preserve the semantic relation between source and target classes when only trained with data from source classes, we address the quantification of the knowledge transfer and semantic relation from an information-theoretic viewpoint. To this end, we follow the prototypical model and format the variables of concern as a probability vector. Leveraging on the proposed probability vector representation, the information measurement such as mutual information and entropy, can be effectively evaluated with simple closed forms. We discuss the choice of common embedding space and distance function when using the prototypical model. Then We propose three information-theoretic loss functions for deterministic GZSL model: a mutual information loss to bridge seen data and target classes; an uncertainty-aware entropy constraint loss to prevent overfitting when using seen data to learn the embedding of target classes; a semantic preserving cross entropy loss to preserve the semantic relation when mapping the semantic representations to the common space. Simulation shows that, as a deterministic model, our proposed method obtains state of the art results on GZSL benchmark datasets. We achieve 21%-64% improvements over the baseline model -- deep calibration network (DCN) and for the first time demonstrate a deterministic model can perform as well as generative ones. Moreover, our proposed model is compatible with generative models. Simulation studies show that by incorporating with f-CLSWGAN, we obtain comparable results compared with advanced generative models.