generalizing results to right rotationally invariant ensembles
- ongoing work for my senior thesis under Pragya Sur and Yufan Li
- we attempt to generalize results from [Hastie et al. '20] to right rotationally invariant design.
TAP free energy for high-dimensional linear regression
- with Viet Vu, under Jiaze Qiu and Subhabrata Sen
- In [Qiu-Sen '22], a TAP equation is proven for for the log-normalizing constant of the posterior distribution in Bayesian linear regression with a uniform spherical prior and i.i.d. Gaussian design, at sufficiently high temperature. We extend these results to all temperatures, under looser conditions on the design. Final project for stat 217.
pruning neural networks at initialization (arxiv)
- with Tanishq Kumar, Mark Sellke
- We prove a modification of the Law of Robustness stated in [Bubeck-Sellke '22] with implications for pruning neural networks at initialization.
witness encryption demo (github, arxiv)
- with Gavin Uberti, Wittmann Goh, Oliver Cheng
- group final project for cs 227