Critical behavior in a cross-situational lexicon learning scenario
This work addresses the problem of understanding early word-learning mechanisms under noisy conditions, with implications for cognitive science and AI, though it is incremental in applying known methods to this specific scenario.
The study investigated how a simple associative learning algorithm acquires word referents in noisy cross-situational scenarios, finding a critical noise threshold beyond which learning becomes impossible and revealing power-law behavior in error-free periods at this threshold.
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 $γ_c$ above which learning is impossible. We use finite-size scaling to show that the sharpness of the transition persists across a region of order $τ^{-1/2}$ about $γ_c$, where $τ$ 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.