Faculty Recruiting Support CICS

Jay-Yoon Lee

Postdoctoral Research Associate


Injecting knowledge/constraints into neural models, primarily for natural language processing tasks. While Machine Learning is a great framework to learn a black-box model that best represents a given training set, sometimes the model fails to learn simple rules that are necessary for safety or logical reasoning.  Jay-Yoon is interested in making the machine learning models more interpretable and logical using human priors on unlabeled datasets.


Jay-Yoon received his Ph.D. in Computer Science from Carnegie Mellon University in 2020, where he was advised by Professor Jaime Carbonell. His thesis focuses on how to inject rules or constraints in the output space into neural NLP models. His research focuses on transfer, multi-task learning where the rules and constraints serve as an anchor to the relationship between domains and tasks.