Thodoris Lykouris

Faculty

Thodoris Lykouris

Support Staff

Title

About

Academic Groups

Academic Area

Thodoris Lykouris is an Assistant Professor of Operations Management at the MIT Sloan School of Management. 

His research focuses on data-driven sequential decision-making and spans across the areas of machine learning, dynamic optimization, and economics. Prior to his current position, he was a postdoctoral researcher at Microsoft Research NYC where he was part of the machine learning group. 

His dissertation was selected as a finalist in the Dantzig dissertation award competition. His papers have also been selected as finalists in the INFORMS Nicholson and Applied Probability Society best student paper competitions. He is also the recipient of a Google Ph.D. Fellowship and a Cornell University Fellowship.

Thodoris holds a Diploma in electrical and computer engineering from National Technical University of Athens (Greece) and a PhD in computer science from Cornell University, where he was advised by Éva Tardos.

 

Publications

"Pricing and Optimization in Shared Vehicle Systems: An Approximation Framework."

Banerjee, Siddhartha, Daniel Freund, and Thodoris Lykouris. Operations Research. Forthcoming. Video. arXiv Preprint.

"Small-loss Bounds for Online Learning with Partial Information."

Lykouris, Thodoris, Karthik Sridharan, and Éva Tardos. Mathematics of Operations Research. Forthcoming. Video.

"Competitive Caching with Machine Learned Advice."

Lykouris, Thodoris and Sergei Vassilvitskii. Journal of the ACM Vol. 68, No. 4 (2021): 1-25. Video.

"Corruption-robust Exploration in Episodic Reinforcement Learning."

Thodoris Lykouris, Max Simchowitz, Aleksandrs Slivkins, and Wen Sun. In In 34th Annual Conference on Learning Theory (COLT 2021), August 2021.

"Contextual Search in the Presence of Irrational Agents."

Akshay Krishnamurthy, Thodoris Lykouris, Chara Podimata, and Robert Schapire. In 53rd Annual ACM Symposium on the Theory of Computing (STOC 2021), New York, NY: June 2021.

"Constrained Episodic Reinforcement Learning in Concave-convex and Knapsack Settings."

Kianté Brantley, Miroslav Dudik, Thodoris Lykouris, Sobhan Miryoosefi, Max Simchowitz, Aleksandrs Slivkins, and Wen Sun. In Proceedings of the 34th Annual Conference on Neural Information Processing Systems (NeurIPS 2020), San Diego, CA: December 2020.

Load More

Recent Insights

Ideas Made to Matter

4 researchers join MIT Sloan faculty

Research topics include analyzing firm behavior, data-driven sequential decision-making, inequality and worker power, and renewable energy integration.

Read Article