CDS PhD Students Co-Author Papers Presented at CogSci 2021 Conference
CDS PhD students Guy Davidson and Yanli Zhou have co-authored papers that were recently presented at the 43rd Annual Meeting of the Cognitive Science Society, CogSci 2021, which took place virtually July 26 — July 29. The conference is a gathering of cognitive scientists, organized by the Vienna Cognitive Science Hub — a network of researchers at the University of Vienna focused in the areas of Cognitive Science and Cognitive Neuroscience.
“Examining Infant Relation Categorization Through Deep Neural Networks” by Guy Davidson and co-author and advisor CDS Assistant Professor Psychology and Data Science, Brenden Lake, models the phenomena of infant relation categorization with deep neural networks. Their findings suggest that these models are useful for studying the computational mechanisms of infant categorization.
“Flexible Compositional Learning of Structured Visual Concepts” by Yanli Zhou (also co-authored and advised by Brenden Lake), studies how people learn different types of visual compositions, using abstract visual forms with rich relational structure. The team develops a Bayesian program induction model that provides a close fit to the behavioral data. The resulting work shows how a single computational approach can account for many distinct types of compositional generalization.
About the Authors
Guy Davidson holds a B.Sc. in Computational Sciences from Minerva University. His research interests center around the intersection between human cognition and machine learning, and particularly, what can we learn from studying humans to allow us to design wiser machine learning systems.
Yanli Zhou holds a BA in Mathematics and Psychology and an MS in Data Science from NYU. Before joining the Lake lab, she worked as a research assistant under the supervision of Dr. Wei Ji Ma at the Center for Neural Science and Department of Psychology at NYU, where she built probabilistic models of visual decision-making tasks. She is broadly interested in incorporating insights from cognitive science into building AI systems that can efficiently and flexibly learn.
By Ashley C. McDonald