I am a PhD student in Computer Science at Brown University, advised by Prof. Randall Balestriero. My work connects the theory and practice of deep learning: theory guides practice, and practice inspires new theory.
Concretely, that runs in two directions.
-
Self-Supervised Learning
When and why representation space is a better place to learn than input space.
-
Spline Theory of Deep Learning
Reading a trained network as a spline operator to describe exactly what it computes.
News
- Jul 2026 “Learning by Reconstruction is an Ill-Defined Prior for Perception” accepted as an extended abstract at TAG-DS 2026.
- Sep 2025 Started my PhD at Brown University with Prof. Randall Balestriero.
- Sep 2025 Curvature Tuning accepted at NeurIPS 2025.
- May 2025 Revolve accepted at ICML 2025.
Selected Publications
The complete and most current list is on Google Scholar.
-
Learning by Reconstruction is an Ill-Defined Prior for Perception: A Level Set View
Encoders with identical reconstruction loss span 4%–74% linear-probe accuracy, so reconstruction underdetermines representation quality — what separates them is optimization’s implicit bias.
-
Is SSL Ready for In-Domain Pretraining? A Cross-Dataset Benchmark and Analysis
A benchmark of six modern SSL methods across 41 image datasets, showing that modern SSL can be competitive with supervised learning for limited-data in-domain pretraining.