I am a fourth year PhD student in theoretical computer science at Carnegie Mellon University, where I am fortunate to be advised by David Woodruff. I graduated with a B.Sc. from Brown University in May of 2017, where I concentrated in mathematics and computer science. In the fall, I will be joining Google NYC as a Research Scientist in the Algorithms and Optimization Group.
My research interests are primarily in sublinear algorithms, especially sketching and streaming algorithms for problems in big-data. In general, I like to think about problems regarding dimensionality reduction; namely, to what extent can we compress the significant components of an enormous, noisy data-set? A common theme of my work is applying sketching techniques to database systems and statistical inference tasks. Additionally, I am interested in property testing, machine learning, and optimization.
An Improved Analysis of the Quadtree for High Dimensional EMD
With Xi Chen, Amit Levi, and Erik Waingarten. [pdf],
[Talk @ CMU Theory Lunch]
Learning and Testing Junta Distributions with Subcube Conditioning
With Xi Chen, Amit Levi, and Erik Waingarten. [arXiv]
Publications
In-Database Regression in Input Sparsity Time
With Alireza Samadian, David Woodruff, and Peng Ye.
ICML 2021
When is Approximate Counting for Conjunctive Queries Tractable?
With Marcelo Arenas, Luis Alberto Croquevielle, and Cristian Riveros.
STOC 2021 [arXiv]
A Framework for Adversarially Robust Streaming Algorithms
With Omri Ben-Eliezer, David Woodruff, and Eylon Yogev.
PODS 2020 [arXiv]PODS Best Paper Award, 2020
Invited to the Journal of the ACM
2021 ACM SIGMOD Research Highlight Award
Invited to HALG 2021
Span Recovery for Deep Neural Networks with Applications to Input Obfuscation
With Qiuyi Zhang and David Woodruff.
ICLR 2020 [arXiv], [Short Talk]
Optimal Sketching for Kronecker Product Regression and Low Rank Approximation
With Huain Diao, Zhao Song, Wen Sun, and David Woodruff.
NeurIPS 2019 [arXiv][Poster]
Towards Optimal Moment Estimation in Streaming and Distributed Models
With David Woodruff.
APPROX 2019 [arXiv]
Learning Two Layer Rectified Neural Networks in Polynomial Time
With Ainesh Bakshi and David Woodruff.
COLT 2019 [arXiv], [Talk @ COLT]
Efficient Logspace Classes for Enumeration, Counting, and Uniform Generation
With Marcelo Arenas, Luis Alberto Croquevielle, and Cristian Riveros.
PODS 2019 [arXiv], [Talk @ CMU Theory Lunch], [SIGMOD Technical Perspective]PODS Best Paper Award, 2019
Invited to the Journal of the ACM
2021 ACM SIGMOD Research Highlight Award
Weighted Reservoir Sampling from Distributed Streams
With Gokarna Sharma, Srikanta Tirthapura, and David P. Woodruff.
PODS 2019 [arXiv]
Perfect L_p Sampling in a Data Stream
With David Woodruff.
FOCS 2018 [arXiv]
Data Streams with Bounded Deletions
With David Woodruff.
PODS 2018 [arXiv]
Approximating Language Edit Distance Beyond Fast Matrix Multiplication: Ultralinear Grammars Are Where Parsing Becomes Hard!
With Barna Saha.
ICALP 2017 [Full Version], [Conference Version]
Miscellaneous:
Learning Stochastically Evolving Networks via Local Probing
Rajesh Jayaram, advised by Eli Upfal (Undergrad Thesis)
[pdf][Defense Slides]
