About me
I am a Postdoc at Columbia University in the Department of Statistics, where I work with Prof. Marco Avella-Medina and Prof. Cynthia Rush. Before joining Columbia in January 2023, I worked as a Senior Researcher at the Robert Bosch Center for Cyber-Physical Systems at the Indian Institute of Science (IISc), Bangalore. I also worked as a Project Scientist at the India Urban Data Exchange program unit at IISc. I completed my PhD from the Department of ECE at IISc in 2021, where I was advised by Prof. Chandra R. Murthy and Prof. Himanshu Tyagi. I work on problems in the areas of high-dimensional statistics, information theory, and privacy.
Preprints
- Lekshmi Ramesh, Elise Han, Marco Avella-Medina, and Cynthia Rush. M-Estimation Under User-Level Local Privacy Constraints. Oct. 2023.
Publications
Anand J. George, Lekshmi Ramesh, Aditya V. Singh, and Himanshu Tyagi. Continual Mean Estimation Under User-Level Privacy. IEEE Journal on Selected Areas in Information Theory, Feb. 2024.
Lekshmi Ramesh, Chandra R. Murthy, and Himanshu Tyagi. Multiple Support Recovery using Very Few Measurements Per Sample. IEEE Transactions on Signal Processing, Apr. 2022.
Full version on arXiv.Lekshmi Ramesh, Chandra R. Murthy, and Himanshu Tyagi. Phase Transitions for Support Recovery from Gaussian Linear Measurements. International Symposium on Information Theory ISIT, Melbourne 2021.
Full version on arXiv.Lekshmi Ramesh, Chandra R. Murthy, and Himanshu Tyagi. Multiple Support Recovery using Very Few Measurements Per Sample. International Symposium on Information Theory ISIT, Melbourne 2021. [Jack Keil Wolf student paper award]
Lekshmi Ramesh, Chandra R. Murthy, and Himanshu Tyagi. Sample-Measurement Tradeoff in Support Recovery Under a Subgaussian Prior. IEEE Transactions on Information Theory, Sept. 2021.
Full version on arXiv.Lekshmi Ramesh, Chandra R. Murthy, and Himanshu Tyagi. Sample-Measurement Tradeoff in Support Recovery Under a Subgaussian Prior. International Symposium on Information Theory ISIT, Paris 2019.
Lekshmi Ramesh and Chandra R. Murthy. Sparse Support Recovery via Covariance Estimation. International Conference on Acoustics, Speech, and Signal Processing ICASSP, Calgary 2018. [Student Paper Award]
Talks
Statistical Inference Under User-Level Local Privacy Constraints. Poster at the North American School of Information Theory, May 2023, University of Pennsylvania and at the Statistical Machine Learning Symposium, Apr. 2023, Columbia University.
Multiple Support Recovery using Very Few Measurements Per Sample. Invited talk at the Berkeley-Columbia Meeting in Engineering and Statistics, Mar. 2023, Columbia University.
Lecture on Concentration Inequalities and Information Theory (YouTube link) for a course on Information Theory offered by Prof. Himanshu Tyagi at CNI, IISc.
Multiple Support Recovery using Very Few Measurements Per Sample, ISIT 2021.
Phase Transitions for Support Recovery from Gaussian Linear Measurements, ISIT 2021.
Support Recovery from Multiple Samples Under Measurement Constraints, Session on Large Scale Statistical Inference at CISS 2021.
Support Recovery from Linear Sketches, CNI Seminar at IISc, August 2020.
Joint Structure Recovery in High Dimensional Data Using Linear Sketches, EECS Research Students Symposium at IISc, July 2020.
Contact
Department of Statistics
SSW Building
1255 Amsterdam Avenue
New York, NY, 10027
Email: lr3160@columbia.edu