Speech Technology and Research (STAR) Laboratory Seminar Series
Upcoming Talks
Abstract: This talk explores and formalizes the view that grammar learning is driven by meaningful language use in context. Taking the human learner as the starting point, I present a computational model of how linguistic constructions (structured mappings between form and meaning) are acquired from pairs of utterances and situation descriptions. The representational basis of the model is a construction-based grammar formalism suitable for capturing constituent structure and relational constraints. This formalism plays a central role in two processes: language understanding, which uses constructions to interpret utterances in context; and language learning, which seeks to improve comprehension by making judicious changes to the current grammar. The resulting integrated model of language structure, use and acquisition offers a more faithful approximation of the child's learning situation than many traditional approaches. Besides offering a concrete realization of a variety of cognitively and developmentally motivated proposals, it also suggests how constraints and assumptions adopted from human language may lead to more effective strategies for dealing with more applied contexts.
Nancy Chang is a research associate at the Université Sorbonne Nouvelle-Paris 3, working with the CoLaJE project on child language acquisition. She earned her doctorate in Computer Science at UC Berkeley, working with the Neural Theory of Language project at the International Computer Science Institute. She has also worked at the Sony Computer Science Laboratory in Paris, focusing on language evolution in populations of robotic agents, and served as a visiting lecturer at Gothenburg University in Sweden. Her research interests center on cognitive and construction-based approaches to building computational models of language structure, learning and use.