Harold R. and Mary Anne Nash Early Career Professor and
Assistant Professor
Education
- Ph.D. Electrical Engineering and Computer Sciences (2020), University of California, Berkeley
- B.Tech. Electrical Engineering (2014), Indian Institute of Technology, Madras
Expertise
- Machine Learning
- Statistics
- Game Theory
- Reinforcement learning
About
Vidya Muthukumar is the Harold R. and Mary Anne Nash Early Career Professor and Assistant Professor in the School of Electrical and Computer Engineering and H. Milton Stewart School of Industrial and Systems Engineering at Georgia Institute of Technology. Dr. Muthukumar received the B.Tech degree from the Indian Institute of Technology, Madras and the Ph.D. degree in Electrical Engineering from University of California, Berkeley. Before joining Georgia Tech, she spent a semester at the Simons Institute for the Theory of Computing as a research fellow for the program “Theory of Reinforcement Learning.” She is the recipient of an Amazon Research Award, NSF CAREER Award, Adobe Data Science Research Award, Simons-Berkeley Google Research Fellowship, and the UC Berkeley EECS Outstanding Course Development and Teaching Award.
Research
Dr. Muthukumar's research interests span a diverse set of topics in the mathematical foundations of machine learning and include:
- online learning
- game theory and game dynamics
- statistical learning theory and high-dimensional statistics
- deep learning theory
- reinforcement learning theory
She is especially interested in designing learning algorithms that provably adapt in strategic environments, fundamental properties of overparameterized models, and the foundations of multi-agent decision-making.
Teaching
Dr. Muthukumar has taught the following courses at Georgia Tech:
- ECE 6756 (earlier 8803): Online Decision Making in Machine Learning - graduate-level course on foundations of online learning, bandits and reinforcement learning that Dr. Muthukumar designed from scratch.
- ISyE 4601 (earlier 4803): Online Learning and Decision Making - undergraduate-level course on foundations of online learning, bandits and reinforcement learning.
- ECE/ISyE/CSE 7750: Mathematical Foundations of Machine Learning - graduate-level course on mathematical foundations of machine learning through the lens of linear algebra, optimization and probability.
She participated in the Class of 1969 Teaching Fellows program in Fall 2021.
Representative Publications
- Guanghui Wang, Krishna Acharya, Lokranjan Lakshmikanthan, Juba Ziani and Vidya Muthukumar: “Last-iterate convergence for symmetric, general-sum, 2×2 games under the exponential weights dynamic”, Algorithmic Learning Theory (ALT), Toronto, 2026 *Best student paper award*
- Kuo-Wei Lai and Vidya Muthukumar: “General Loss Functions Lead to (Approximate) Interpolation in High Dimensions”, Journal of Machine Learning Research, 2025 issue.
- Guanghui Wang, Zihao Hu, Claudio Gentile, Vidya Muthukumar and Jacob Abernethy: “Faster margin maximization rates for generic and adversarially robust optimization methods”, Mathematical Programming Series B (special issue on Optimization for Machine Learning), 2025 issue.
- Milind Nakul, Vidya Muthukumar and Ashwin Pananjady: “Estimating stationary mass, frequency by frequency”, Conference on Learning Theory (COLT), Lyon, 2025.
- Vidya Muthukumar, Soham Phade and Anant Sahai: “On the impossibility of convergence of strategies arising from no-regret learning”, Mathematics of Operations Research, 2024 issue.
- Chi-Heng Lin, Chiraag Kaushik, Eva L. Dyer and Vidya Muthukumar: “The good, bad and ugly sides of data augmentation: An implicit spectral regularization perspective”, Journal of Machine Learning Research, 2024 issue.
- Tyler LaBonte, Vidya Muthukumar and Abhishek Kumar: “Towards last-layer retraining for group robustness with fewer annotations”, Neural Information Processing Systems (NeurIPS), New Orleans, 2023.
- Yujia Jin, Vidya Muthukumar and Aaron Sidford: “The complexity of infinite-horizon general-sum stochastic games”, Innovations in Theoretical Computer Science (ITCS) 2023.
- Vidya Muthukumar, Adhyyan Narang, Vignesh Subramanian, Mikhail Belkin, Daniel Hsu and Anant Sahai: “Classification vs regression in overparameterized regimes: Does the loss function matter?”, Journal of Machine Learning Research, 2021 issue.
- Vidya Muthukumar, Kailas Vodrahalli, Vignesh Subramanian and Anant Sahai: “Harmless interpolation of noisy data in linear regression”, IEEE Journal of Selected Areas in Information Theory, inaugural special issue on “Deep Learning: Mathematical Foundations and Applications to Information Science”, 2020 issue.
- Vidya Muthukumar, Mitas Ray, Anant Sahai and Peter L. Bartlett: “Best of many worlds: Robust model selection for online supervised learning”, International Conference on Artificial Intelligence and Statistics (AISTATS), 2019, Naha, Okinawa.