The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
Jonathan Frankle is a PhD student in MIT’s Department of Electrical Engineering and Computer Science, where he studies empirical deep learning. His current research focus is on the properties of sparse networks that allow them to train effectively as embodied by his “Lottery Ticket Hypothesis” which received a best paper award at ICLR 2019. He also works closely with lawyers, journalists, and policymakers on topics in AI policy and has taught at the Georgetown University Law Center. He earned a BSE and MSE in computer science at Princeton University.
- Frankle, J., and Carbin, M. (2019) The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks. International Conference on Learning Representations (ICLR).
- Frankle, J., Dziugaite, G., Roy, D., and Carbin, M. (2020) Linear Mode Connectivity and the Lottery Ticket Hypothesis. International Conference on Machine Learning (ICML).
Garvie, C., Bedoya, A., and Frankle, J. (2016) The Perpetual Line-Up: Unregulated Police Face Recognition in America. Investigative Report.
- May 10, 2019. MIT Technology Review. A new way to build tiny neural networks could create powerful AI on your phone.
- Apr. 14, 2019. New York Times. One Month, 500,000 Face Scans: How China Is Using A.I. to Profile a Minority.
- Aug. 16, 2018. New York Times. An Airline Scans Your Face. You Take Off. But Few Rules Govern Where Your Data Goes.
Learning Rate Rewinding for elegant neural network pruning
Publications with the MIT-IBM Watson AI Lab