Rebecca Roelofs

I am a fifth year Ph.D. student in Computer Science at U.C. Berkeley. I am coadvised by Benjamin Recht and Jim Demmel.

In 2016, I was an intern at Google Brain in NYC, working with Vikas Sindhwani.

I graduated from Swarthmore College in 2013 with a B.A. in Computer Science and a B.S.E. in Engineering. While there, I conducted research with Matt Zucker and Tali Moreshet.

My research interests include first order optimization methods for machine learning, generalization, large scale iterative solvers, and numerical linear algebra.


Wilson, A. C., Roelofs, R., Stern, M., Srebro, N., & Recht, B. "The Marginal Value of Adaptive Gradient Methods in Machine Learning." NIPS, 2017. [paper, slides, poster]

Sindhwani, Vikas, Rebecca Roelofs, and Mrinal Kalakrishnan. "Sequential operator splitting for constrained nonlinear optimal control." American Control Conference (ACC), 2017. IEEE, 2017.

Tu, S., Roelofs, R., Venkataraman, S., & Recht, B. (2016). Large scale kernel learning using block coordinate descent. arXiv preprint arXiv:1602.05310.

Stromme, A., Sutherland, D. J., Felt, N., Burka, A., Lipton, B., Roelofs, R., ... & Welkie, A. (2012, December). Managing User Requests With the Grand Unified Task System (GUTS). In LISA (pp. 101-110).


In Spring 2016, I was a TA for EE227C: Convex Optimization for Modern Data Analysis taught by Benjamin Recht.

In Fall 2016, I was the head TA for CS188: Introduction to Artificial Intelligence taught by Josh Hug.