Ross Boczar, PhD

rjboczar A-T gmail D-O-T com (Does this do anything?)

Bio

My research interests include control, learning theory, optimization, and other areas of applied mathematics.

I am currently an IFDS Postdoctoral Fellow at the University of Washington, working with Prof. Kevin Jamieson and Prof. Lillian Ratliff.

Previously:

I worked at Amazon Advertising and Amazon Go as an Applied Scientist from 2019-2021.
In May 2019, I received my PhD in Electrical Engineering and Computer Sciences from UC Berkeley. While at Berkeley, I was advised by Prof. Ben Recht in association with BAIR and was supported by the DoD NDSEG Fellowship.
In 2016, I was an intern at Google Seattle.
In 2015, I was a GSI for the first major-agnostic introductory data science course. During the course, I spent some time cramming machine learning classifiers into Tweets.
I worked in the RISE Lab at NAVAIR in Lakehurst, NJ from 2012-2014,.
In 2012, I received my MS in Electrical Engineering along with a BS in Electrical Engineering and a minor in Mathematics from the University of Pennsylvania.

I enjoy basketball and music.

Publications

Preprints

Arnab Maiti*, Ross Boczar*, Kevin Jamieson, Lillian J. Ratliff. "Near-Optimal Pure Exploration in Matrix Games: A Generalization of Stochastic Bandits & Dueling Bandits." [arXiv]
Arnab Maiti, Ross Boczar, Kevin Jamieson, Lillian J. Ratliff. "Query-Efficient Algorithms to Find Unique Nash Equilibria in Two-Player Zero-Sum Matrix Games." [arXiv]
"Fair Participation via Sequential Policies."
Stephen Tu, Ross Boczar, "An elementary proof of anti-concentration for degree two non-negative Gaussian polynomials." [arXiv]
Stephen Tu, Ross Boczar, Andrew Packard, Benjamin Recht, "Non-Asymptotic Analysis of Robust Control from Coarse-Grained Identification." [arXiv]
Ross Boczar, Laurent Lessard, Andrew Packard, Benjamin Recht, "Exponential Stability Analysis via Integral Quadratic Constraints." Submitted to IEEE Transactions on Automatic Control. [arXiv]

Conference Proceedings

Stephen Tu*, Ross Boczar*, Benjamin Recht, "Minimax Lower Bounds for H-infinity-Norm Estimation." American Control Conference 2019. [arXiv]
Max Simchowitz* , Ross Boczar*, Benjamin Recht, "Learning a Marginally Stable System from One Trajectory." Conference on Learning Theory 2019. [arXiv]
Ross Boczar, Nikolai Matni, Benjamin Recht, "Finite-Data Performance Guarantees for the Output-Feedback Control of an Unknown System." IEEE Conference on Decision and Control 2018. [arXiv]
Stephen Tu, Ross Boczar, Benjamin Recht, "On the Approximation of Toeplitz Operators for Nonparametric H-infinity-norm Estimation." American Control Conference 2018. [arXiv]
Stephen Tu, Ross Boczar, Max Simchowitz, Mahdi Soltanolkotabi, Benjamin Recht, "Low-rank Solutions of Linear Matrix Equations via Procrustes Flow." ICML 2016. [Link][arXiv]
Ross Boczar, Laurent Lessard, Benjamin Recht, "Exponential Convergence Bounds using Integral Quadratic Constraints." 54th IEEE Conference on Decision and Control 2015, Osaka, Japan. [Link][arXiv][Code]
James T. Hing, Ross Boczar, Kyle Hart, "Parking Autonomous Skids." Human Interface and the Management of Information: Information and Knowledge in Context. 17th International Conference, HCI International 2015, Los Angeles, CA, USA, August 2-7, 2015, Proceedings, Part II. [Link]
Larry Venetsky, Ross Boczar, and Robert Lee-Own, "Optimization of background subtraction for image enhancement," Proc. SPIE 8751, Machine Intelligence and Bio-inspired Computation: Theory and Applications VII, 875102 (May 28, 2013), doi:10.1117/12.2014319. [PDF]

* Equal contribution.

Technical Reports

Ross Boczar, "Performance Guarantees in Learning and Robust Control." PhD Dissertation, 2019. [Link]
Ross Boczar, Jason Suapengco, and Gabriel Torres, "AutoPlug: Open Architecture for Automotive Services," 2011. [Poster]

Other

Amy Calhoun, "AutoPlug: Plug-n-Play Comes to Automotive Technology," Penn Engineering Magazine, Fall 2011. [PDF]
Some patents...
Some talks...