Probabilistic AND-OR Attribute Grouping for Zero-Shot Learning

TL;DR

This paper introduces LAGO, a probabilistic model that captures attribute group structures with soft and-or relations, improving zero-shot learning accuracy and unifying existing approaches like DAP and ESZSL.

Contribution

LAGO is a novel probabilistic framework that learns attribute groupings and their relations end-to-end, enhancing zero-shot learning performance and providing a unified formulation of prior methods.

Findings

LAGO improves zero-shot learning accuracy on two of three benchmarks.

Soft attribute groupings learned by LAGO capture meaningful structures.

A relaxed DAP approach based on LAGO outperforms original DAP by ~40%.

Abstract

In zero-shot learning (ZSL), a classifier is trained to recognize visual classes without any image samples. Instead, it is given semantic information about the class, like a textual description or a set of attributes. Learning from attributes could benefit from explicitly modeling structure of the attribute space. Unfortunately, learning of general structure from empirical samples is hard with typical dataset sizes. Here we describe LAGO, a probabilistic model designed to capture natural soft and-or relations across groups of attributes. We show how this model can be learned end-to-end with a deep attribute-detection model. The soft group structure can be learned from data jointly as part of the model, and can also readily incorporate prior knowledge about groups if available. The soft and-or structure succeeds to capture meaningful and predictive structures, improving the accuracy of zero-shot learning on two of three benchmarks. Finally, LAGO reveals a unified formulation over two ZSL approaches: DAP (Lampert et al., 2009) and ESZSL (Romera-Paredes & Torr, 2015). Interestingly, taking only one singleton group for each attribute, introduces a new soft-relaxation of DAP, that outperforms DAP by ~40.