Randall Balestriero
About
Randall Balestriero is an Assistant Professor of Computer Science at Brown University, where he leads the Balestriero Lab. His research is dedicated to closing the gap between the empirical success of deep learning and its theoretical foundations. He is widely recognized for his work on Self-Supervised Learning (SSL) and for developing a "unified theory" of deep networks using spline theory. By viewing neural networks as collections of local linear models (affine splines), he provides geometric and mathematical insights into how models like Large Language Models (LLMs) and Joint Embedding Predictive Architectures (JEPA) represent data. Before joining Brown in 2024, he was a postdoctoral researcher at Meta AI (FAIR) working with Yann LeCun, and he previously held research roles at Citadel and GQS.
Research Performance Summary
First Recorded Paper
Gabor scalogram extracts dolphin click formants
Year: 2013
Citations: 1
Venue: Proc. 1st workshop Neural Information Processing Scaled for Bioacoustics
Latest Recorded Paper
Multi-scale clustering and source separation of InSight mission seismic data
Year: 2026
Citations: 0
Venue: IEEE Transactions on Geoscience and Remote Sensing
Last 10 Years Publication Activity
This timeline shows the professor's yearly publication activity.