Critical behavior in a cross-situational lexicon learning scenario

TL;DR

This paper investigates how noise affects associative learning in cross-situational word learning, identifying a critical noise level where learning fails and analyzing the transition's sharpness and duration distributions.

Contribution

It introduces a critical noise threshold in a simple associative model and characterizes the phase transition and error scaling near this critical point.

Findings

Existence of a critical noise parameter $oldsymbol{\gamma_c}$ for learning failure

Sharp transition in learning performance around $oldsymbol{\gamma_c}$

Power-law distribution of learning error-free periods at criticality

Abstract

The associationist account for early word-learning is based on the co-occurrence between objects and words. Here we examine the performance of a simple associative learning algorithm for acquiring the referents of words in a cross-situational scenario affected by noise produced by out-of-context words. We find a critical value of the noise parameter $\gamma_c$ above which learning is impossible. We use finite-size scaling to show that the sharpness of the transition persists across a region of order $\tau^{-1/2}$ about $\gamma_c$, where $\tau$ is the number of learning trials, as well as to obtain the learning error (scaling function) in the critical region. In addition, we show that the distribution of durations of periods when the learning error is zero is a power law with exponent -3/2 at the critical point.