A Bayesian generative model for learning semantic hierarchies.

Roni Mittelman,Benjamin Kuipers,Min Sun,Silvio Savarese

doi:10.3389/fpsyg.2014.00417

Roni Mittelman, Benjamin Kuipers + Show 2 more

Open Access

https://doi.org/10.3389/fpsyg.2014.00417

Copy DOI

Abstract

Building fine-grained visual recognition systems that are capable of recognizing tens of thousands of categories, has received much attention in recent years. The well known semantic hierarchical structure of categories and concepts, has been shown to provide a key prior which allows for optimal predictions. The hierarchical organization of various domains and concepts has been subject to extensive research, and led to the development of the WordNet domains hierarchy (Fellbaum, 1998), which was also used to organize the images in the ImageNet (Deng et al., 2009) dataset, in which the category count approaches the human capacity. Still, for the human visual system, the form of the hierarchy must be discovered with minimal use of supervision or innate knowledge. In this work, we propose a new Bayesian generative model for learning such domain hierarchies, based on semantic input. Our model is motivated by the super-subordinate organization of domain labels and concepts that characterizes WordNet, and accounts for several important challenges: maintaining context information when progressing deeper into the hierarchy, learning a coherent semantic concept for each node, and modeling uncertainty in the perception process.

Highlights

A Bayesian generative model for learning semantic hierarchiesThe hierarchical organization of various domains and concepts has been subject to extensive research, and led to the development of the WordNet domains hierarchy (Fellbaum, 1998), which was used to organize the images in the ImageNet (Deng et al, 2009) dataset, in which the category count approaches the human capacity
There has been mounting evidence in recent years for the role that Bayesian probabilistic computations play both in the behavioral and the neural circuit layers of cognition (Chater et al, 2006; Steyvers et al, 2006; Tenenbaum et al, 2006; Fiser et al, 2010)
Bayesian models have become an important tool for describing cognitive processes, and we propose a Bayesian generative model that learns a semantic hierarchy based on observations of objects in a concept space in which objects are represented as binary attribute vectors

Summary

A Bayesian generative model for learning semantic hierarchies

The hierarchical organization of various domains and concepts has been subject to extensive research, and led to the development of the WordNet domains hierarchy (Fellbaum, 1998), which was used to organize the images in the ImageNet (Deng et al, 2009) dataset, in which the category count approaches the human capacity. We propose a new Bayesian generative model for learning such domain hierarchies, based on semantic input. Our model is motivated by the super-subordinate organization of domain labels and concepts that characterizes WordNet, and accounts for several important challenges: maintaining context information when progressing deeper into the hierarchy, learning a coherent semantic concept for each node, and modeling uncertainty in the perception process

INTRODUCTION

WORDNET DOMAINS HIERARCHY

A BAYESIAN GENERATIVE MODEL FOR LEARNING DOMAIN HIERARCHIES

EXPERIMENTS

EVALUATING THE GENERATIVE MODEL AS A HIERARCHICAL CLUSTERING ALGORITHM

DISCUSSION