Human Inspired Progressive Alignment and Comparative Learning for Grounded Word Acquisition

TL;DR

This paper introduces a human-inspired progressive learning method for grounded word acquisition, enabling models to learn language concepts continually by comparing attribute similarities without fixed vocabularies.

Contribution

It presents a novel comparative learning framework inspired by human language acquisition, allowing continual and efficient grounded word learning without fixed vocabulary constraints.

Findings

Effective continual learning of grounded words demonstrated

Models can filter common information for shared labels

Approach aligns with cognitive findings on language acquisition

Abstract

Human language acquisition is an efficient, supervised, and continual process. In this work, we took inspiration from how human babies acquire their first language, and developed a computational process for word acquisition through comparative learning. Motivated by cognitive findings, we generated a small dataset that enables the computation models to compare the similarities and differences of various attributes, learn to filter out and extract the common information for each shared linguistic label. We frame the acquisition of words as not only the information filtration process, but also as representation-symbol mapping. This procedure does not involve a fixed vocabulary size, nor a discriminative objective, and allows the models to continually learn more concepts efficiently. Our results in controlled experiments have shown the potential of this approach for efficient continual learning of grounded words.