In this project, we plan to develop and implement a computational model for how children acquire a language that is grounded in contemporary theories of human language syntax known as minimalism. The project would include curating and analyzing data from the CHILDES corpus of mother-child speech and formulating hypotheses about the developmental trajectories of children. We will probe several broad learning acquisition hypotheses and examine how structural and statistical principles can be traded off against each other in child acquisition. In the broader sense, this can contribute to the current linguistic theories regarding child language acquisition and statistical inference with small data as well as bringing a unique computational approach to bear on this problem.

Through this SuperUROP, I wish to learn new skills and enhance my knowledge of cutting-edge research at the confluence of my two majors, computer science and linguistics. I have worked on research projects in pure linguistics, machine learning and natural language processing, and I am excited for the opportunity to bring my interests and skills together. I hope to carry out a successful research project with meaningful results to publish.