- This event has passed.
FOLDS seminar: Multi-step reasoning via curriculum learning
February 26 at 12:00 PM - 1:00 PM
Zoom link: https://upenn.zoom.us/j/
Can multi-step reasoning be learned from data? We investigate this question in the context of a simple function composition task. We prove that this task is hard to learn in the Statistical Query model, but is easy to learn with transformers under various forms of curriculum learning. This is joint work with Zixuan Wang, Eshaan Nichani, Alberto Bietti, Alex Damian, Jason Lee, and Denny Wu.
Daniel Hsu
Associate professor in the Department of Computer Science and Associate Director for Research for the Data Science Institute, at Columbia University
Daniel Hsu is an associate professor in the Department of Computer Science and the Associate Director for Research for the Data Science Institute, both at Columbia University. He works on algorithmic statistics and machine learning, with the goals of designing efficient algorithms for learning and data analysis, and understanding the limits of efficient computation for these tasks. Daniel completed his Ph.D. at UC San Diego and his B.S. at UC Berkeley. He was a postdoc at the Departments of Statistics at Rutgers University and the University of Pennsylvania and also at Microsoft Research New England. He was selected by IEEE Intelligent Systems as one of “AI’s 10 to Watch” in 2015 and received a Sloan Research Fellowship in 2016. He has served as a program chair for the Conference on Learning Theory, the International Conference on Algorithmic Learning Theory, and the International Conference on Machine Learning.
His Ph.D. advisor at UCSD was the glorious Sanjoy Dasgupta. His postdoctoral stints at Penn and Rutgers were with the equally glorious Sham Kakade and Tong Zhang.