Computational Models of Language Acquisition

Abstract

Abstract. Child language acquisition, one of Nature's most fascinating phenomena, is to a large extent still a puzzle. Experimental evidence seems to support the view that early language is highly formulaic, consisting for the most part of frozen items with limited productivity. Fairly quickly, however, children find patterns in the ambient language and generalize them to larger structures, in a process that is not yet well understood. Computational models of language acquisition can shed interesting light on this process. This paper surveys various works that address language learning from data; such works are conducted in different fields, including psycholinguistics, cognitive science and computer science, and we maintain that knowledge from all these domains must be consolidated in order for a well-informed model to emerge. We identify the commonalities and differences between the various existing approaches to language learning, and specify desiderata for future research that must be considered by any plausible solution to this puzzle.

On the Trade-off between Redundancy and Local Coherence in Summarization

“…Firstly, eye tracking and brain activity data (captured by functional magnetic resonance imaging, fMRI, and electroencephalography, EEG) proved useful for a wide range of tasks such as sentiment analysis (Gu et al, 2014;Mishra et al, 2018), relation extraction (McGuire & Tomuro, 2021), name entity recognition , and text simplification (Klerke et al, 2016). Secondly, cognitive theories of text comprehension and production have guided model design for grammar induction and constituency parsing (Levy et al, 2008;Wintner, 2010), machine translation (Saini & Sahula, 2021), common-sense reasoning (Sap et al, 2020), and training strategies involving regularization (Wei et al, 2021) and curriculum learning (Xu et al, 2020). Moreover, these theories played key roles in the understanding catastrophic forgetting during fine-tuning of neural networks (Arora, Rahimi, & Baldwin, 2019), better analysis of what neural networks comprehend (Ettinger, 2020;Dunietz et al, 2020), and better evaluation of generated text (van Der Lee et al, 2019),…”

Representing and Using Knowledge with the Contextual Evaluation Model

“…The language learning process presented in this paper differs from many of the existing models [37,38] in several respects. First is that it assumes no a priori knowledge and no external 'guidance'.…”

Aprendizagem de categorias de palavras por análise distribucional resultados adicionais para Português Brasileiro

“…Given the increasing number of modeling studies being conducted nowadays, it becomes more and more important to be able to assess their replicability, a goal for which this paper makes a humble contribution. Finally, although related works have appeared before and many did since then, Redington et al's (1998) study isto our knowledge (Frank 2011;Kaplan, Oudeyer & Bergen 2008;Seidenberg 1997;Wintner 2010;Yang 2012) and in this specific subjectthe first and most comprehensive computational study on the distributional properties of child directed speech and how it relates to language acquisition. In this regard, this model is a computational cognitive model.…”