A Simple Exponential Family Framework for Zero-Shot Learning

TL;DR

This paper introduces a straightforward generative exponential family framework for zero-shot learning that models classes as probability distributions, enabling effective prediction of unseen classes and easy extension to few-shot learning.

Contribution

It proposes a novel exponential family-based generative model that predicts class distributions from attributes, improving zero-shot and few-shot learning capabilities.

Findings

Effective on multiple benchmark datasets

Allows semi-supervised and transductive learning

Seamless extension to few-shot learning

Abstract

We present a simple generative framework for learning to predict previously unseen classes, based on estimating class-attribute-gated class-conditional distributions. We model each class-conditional distribution as an exponential family distribution and the parameters of the distribution of each seen/unseen class are defined as functions of the respective observed class attributes. These functions can be learned using only the seen class data and can be used to predict the parameters of the class-conditional distribution of each unseen class. Unlike most existing methods for zero-shot learning that represent classes as fixed embeddings in some vector space, our generative model naturally represents each class as a probability distribution. It is simple to implement and also allows leveraging additional unlabeled data from unseen classes to improve the estimates of their class-conditional distributions using transductive/semi-supervised learning. Moreover, it extends seamlessly to few-shot learning by easily updating these distributions when provided with a small number of additional labelled examples from unseen classes. Through a comprehensive set of experiments on several benchmark data sets, we demonstrate the efficacy of our framework.