I am a fourth-year Ph.D. candidate at MIT-EECS, advised by Professor Ashia Wilson.
My academic journey began with a B.Sc. degree in EE from the Technion - Israel Institute of Technology, followed by an M.Sc. degree in EE from Tel-Aviv University, where I had the privilege of being advised by Professor Anatoly Khina.
My research interests lie at the intersection of machine learning, differential privacy, information theory, signal processing, and wireless communications. I am particularly interested in developing mathematically grounded engineering methods for addressing important real-world challenges.
My current research interests are:
Randomized sketching for differential privacy and optimization
Machine learning for wireless communications and signal processing
Classical topics in information theory
Email: omrilev@mit.edu
Omri Lev, Moshe Shenfeld, Vishwak Srinivasan, Katrina Ligett, and Ashia C. Wilson, ``Near-Optimal Private Linear Regression via Iterative Hessian Mixing", arXiv; Accepted for publication on the Forty-Third International Conference on Machine Learning (ICML), spotlight; Early version accepted at TPDP 2026;
Omri Lev, Vishwak Srinivasan*, Moshe Shenfeld*, Katrina Ligett, Ayush Sekhari, and Ashia C. Wilson, ``The Gaussian Mixing Mechanism: Rényi Differential Privacy via Gaussian Sketches", The conference on Neural Information Processing Systems (NeurIPS); Early version presented at TPDP 2025; arXiv
Omri Lev and Ashia C. Wilson, ``Influence Estimation in Statistical Models Using the Fisher Information Matrix”, Transactions on Machine Learning Research (TMLR). Link. Earlier version presented at the MIT Robustness and Influence Functions Workshop;
Omri Lev and Anatoly Khina ``Energy-limited Joint Source-Channel Coding of Gaussian Sources over Gaussian Channels with Unknown Noise Level”. Entropy: Advances in Information and Coding Theory II, Link;
Omri Lev and Anatoly Khina ``Energy-limited Joint Source–Channel Coding via Analog Pulse Position Modulation”. IEEE Transactions on Communications, Volume: 70, Issue: 8. Link;
Omri Lev and Amir Weiss, ``Near-Optimal Online Gain Control for Modulo ADCs”, Accepted for publication at the International Conference on Acoustics, Speech and Signal Processing (ICASSP), May. 2025, Link;
Omri Lev, Matthew Ho, Ligong Wang, and Gregory W. Wornell, ``On the Capacity of Scalar Gaussian Channels Subject to State Obfuscation”, 60th Allerton Conference, September. 2024, Link;
Omri Lev and Anatoly Khina, ``Universal Joint Source-Channel Coding Under an Input Energy Constraint”, The 2022 IEEE International Symposium on Information Theory (ISIT), Jun. 2022, Link;
Omri Lev and Anatoly Khina, ``Schemes for LQG Control Over Gaussian Channels With Side Information”, XVII International Symposium for Problems of Redundancy in Information and Control Systems, Oct. 2021, Link;
Omri Lev and Anatoly Khina, ``Energy-Limited Joint Source–Channel Coding via Analog Pulse Position Modulation”, Proceedings of the IEEE Information Theory Workshop (ITW), Oct. 2021, Link;
Omri Lev and Anatoly Khina, ``Gauss–Markov Source Tracking With Side Information: Lower Bounds", International Symposium on Information Theory and Its Applications (ISITA), Oct. 2020, Link;
Omri Lev, Tal Wiener, Deborah Cohen, and Yonina C. Eldar, ``WINDOW: Wideband Demodulator for Optical Waveforms", Optics Express, Vol:25, Pages: 19444-19456, Link;