Back to homepage

Research

My interests broadly lie in machine learning and algorithm design. My current work is in topics in statisical learning theory, including sample compression and online learning. I have also done work in the past in optimal transport.

Preprints

Publications

2025

Attias, I., Hanneke, S., and Ramaswami, A. (2025). Sample Compression Scheme Reductions. In Proceedings of the 36th International Conference on Algorithmic Learning Theory (ALT). arxiv

Tradeoffs between Mistakes and ERM Oracle Calls in Online and Transductive Online Learning. Idan Attias, Steve Hanneke, Arvind Ramaswami. Spotlight paper at 39th Annual Conference on Neural Information Processing Systems (NeurIPS 2025). arxiv

2023

Chen, J., Chen, L., Liu, Y. P., Peng, R., & Ramaswami, A. (2023). Exponential Convergence of Sinkhorn Under Regularization Scheduling. In SIAM Conference on Applied and Computational Discrete Algorithms (ACDA23) (pp. 180-188). Society for Industrial and Applied Mathematics. arxiv, Proceedings, SIAM ACDA 2023

Other writeups (Class projects)

SDPs for MaxCut Approximations (2021, Final project in MATH 7014 - Advanced Graph Theory). (Survey, Presentation)

SQ Learning for Tensor PCA (2021, Research report in CS7545 - Machine Learning Theory) (Survey)