Word learning and the acquisition of syntactic-semantic overhypotheses

Children learning their first language face multiple problems of induction: how to learn the meanings of words, and how to build meaningful phrases from those words according to syntactic rules. We consider how children might solve these problems efficiently by solving them jointly, via a computational model that learns the syntax and semantics of multi-word utterances in a grounded reference game. We select a well-studied empirical case in which children are aware of patterns linking the syntactic and semantic properties of words --- that the properties picked out by base nouns tend to be related to shape, while prenominal adjectives tend to refer to other properties such as color. We show that children applying such inductive biases are accurately reflecting the statistics of child-directed speech, and that inducing similar biases in our computational model captures children's behavior in a classic adjective learning experiment. Our model incorporating such biases also demonstrates a clear data efficiency in learning, relative to a baseline model that learns without forming syntax-sensitive overhypotheses of word meaning. Thus solving a more complex joint inference problem may make the full problem of language acquisition easier, not harder.

[1]  Afsaneh Fazly,et al.  A Probabilistic Computational Model of Cross-Situational Word Learning , 2010, Cogn. Sci..

[2]  Mark T. Greenberg,et al.  Environmental influences on early language development: The context of social risk , 1990, Development and Psychopathology.

[3]  Linda B. Smith,et al.  Object name Learning Provides On-the-Job Training for Attention , 2002, Psychological science.

[4]  Matthias Scheutz,et al.  Early Syntactic Bootstrapping in an Incremental Memory-Limited Word Learner , 2018, AAAI.

[5]  J. Tenenbaum,et al.  The learnability of abstract syntactic principles , 2011, Cognition.

[6]  Susan A. Gelman,et al.  Adjectives and Nouns: Children's Strategies for Learning New Words. , 1988 .

[7]  Toben H. Mintz,et al.  Adjectives really do modify nouns: the incremental and restricted nature of early adjective acquisition , 2002, Cognition.

[8]  J. Tenenbaum,et al.  Bayesian Special Section Learning Overhypotheses with Hierarchical Bayesian Models , 2022 .

[9]  Martha Jo-Ann. Demetras,et al.  WORKING PARENTS' CONVERSATIONAL RESPONSES TO THEIR TWO-YEAR-OLD SONS (LINGUISTIC INPUT, LANGUAGE ACQUISITION). , 1986 .

[10]  K. Nelson Narratives from the crib , 1990 .

[11]  Linda B. Smith,et al.  Early noun vocabularies: do ontology, category structure and syntax correspond? , 1999, Cognition.

[12]  Grzegorz Chrupala,et al.  Learning language through pictures , 2015, ACL.

[13]  Alon Lavie,et al.  High-accuracy Annotation and Parsing of CHILDES Transcripts , 2007 .

[14]  E. Clark Awareness of Language: Some Evidence from what Children Say and Do , 1978 .

[15]  Luke S. Zettlemoyer,et al.  Weakly Supervised Learning of Semantic Parsers for Mapping Instructions to Actions , 2013, TACL.

[16]  Luke S. Zettlemoyer,et al.  Online Learning of Relaxed CCG Grammars for Parsing to Logical Form , 2007, EMNLP.

[17]  Nathaniel J. Smith,et al.  Bootstrapping language acquisition , 2017, Cognition.

[18]  Barbara Landau,et al.  Count nouns, adjectives, and perceptual properties in children's novel word interpretations. , 1992 .

[19]  Chen Yu,et al.  On the Integration of Grounding Language and Learning Objects , 2004, AAAI.

[20]  Mark Steedman,et al.  Surface structure and interpretation , 1996, Linguistic inquiry.

[21]  J. Bohannon,et al.  Children's control of adult speech. , 1977 .