A Concept Learning Approach to Multisensory Object Perception

TL;DR

This paper introduces a Bayesian concept learning model for multisensory object recognition, integrating probabilistic grammar and inference to improve understanding of how humans learn and recognize complex 3D objects across senses.

Contribution

It presents a novel computational model that combines probabilistic grammar with Bayesian inference for multisensory concept learning of complex objects.

Findings

Model successfully recognizes multisensory 3D objects

Integrates visual and haptic data effectively

Advances understanding of concept learning mechanisms

Abstract

This paper presents a computational model of concept learning using Bayesian inference for a grammatically structured hypothesis space, and test the model on multisensory (visual and haptics) recognition of 3D objects. The study is performed on a set of artificially generated 3D objects known as fribbles, which are complex, multipart objects with categorical structures. The goal of this work is to develop a working multisensory representational model that integrates major themes on concepts and concepts learning from the cognitive science literature. The model combines the representational power of a probabilistic generative grammar with the inferential power of Bayesian induction.