
Trapit Bansal
University of Massachusets Amherst
1
presentations
1
number of views
SHORT BIO
Trapit is a Ph.D. student advised by Prof. Andrew McCallum at UMass Amherst. His recent research focuses on improving the generalization of natural language processing models with limited human-labeled data through meta-learning, self-supervised learning, and multi-task learning. In the past, he has also worked on machine learning methods for recommendation systems, information extraction, knowledge representation, and reinforcement learning for multi-agent systems. During his Ph.D., he has interned at Facebook, OpenAI, Google Research, and Microsoft Research. His work has also received a best paper award at ICLR 2018. Before starting his Ph.D., he obtained a B.S. and M.S. in Mathematics from the Indian Institute of Technology, Kanpur.
Presentations

Diverse Distributions of Self-Supervised Tasks for Meta-Learning in NLP
Trapit Bansal