Hi, Thanks for your interest!
I am an assistant professor at the Department of Computer Science, University of Illinois (UIC), Chicago. I belong to the UIC CS Theory group, and a member of the IDEAL Institute. Before this, I was a research scientist at Apple MLR, broadly working in the area of Machine Learning theory and a short-term visiting-faculty at Toyota Technological Institute at Chicago (TTIC). I completed my postdoc at Microsoft Research (NYC) and PhD from the department of Computer Science, Indian Institute of Science, Bangalore. I was fortunate to be advised by Aditya Gopalan and Chiranjib Bhattacharyya during my PhD, and to have been a part of several internships, including Microsoft Research (In), Inria (Paris), and Google AI (MTV).
Main Research Focus: AI-Alignment with RLHF.Research Areas: Machine Learning (esp. Online Learning theory, Bandits, Reinforcement Learning), Optimization, Federated Learning, Differential Privacy, and Mechanism Design.
Brief Research Summary
My primary research focuses on developing, improving, and theoretically analyzing reinforcement learning with human feedback (RLHF) algorithms to enhance machine assistance for humanity. Applications include language models, assistive robotics, autonomous driving, and personalized systems—-almost any system that can improve itself through user interaction.
A bit more generally, my work focuses on building large-scale, robust, and intelligent AI models for sequential decision-making under partial or restricted feedback, such as user interactions, preferences, demonstrations, proxy observations, and rankings. In the past, I have also explored combinatorial decision spaces, dynamic regret, multiplayer games, and distributed optimization. My work falls under various interdisciplinary research areas, including ML, learning theory, optimization, operations research, mechanism design, privacy, federated learning, and algorithmic fairness. Feel free to reach out if you are interested in exploring related topics!
[Selected Papers] [Full List] [Google Scholar] [DBLP] [arXiv]
Prospective Students: I am looking to work with ML enthusiasts and motivated students! If you are interested in joining our research group, please send me your resume and research statement. I also encourage you to apply to the UIC-CS PhD program (and do mention my name in your application).
On a personal note
I love my father's lectures and am a proud daughter of Prof. Saha (as fondly popular among his students, being a super strict teacher yet a truly charismatic and caring mentor)! I deeply wish to contribute to the education of students battling with Hemophilia. If you are connected to any Hemophilia Welfare Organization and believe that my support in any capacity can make a difference, please do not hesitate to reach out. Carrying forward my father's passion, I would love to collaborate with any organization, research lab, or individual who shares the mission of improving the lives of hemophilic students and supporting their education.
Selected Papers:
- Stop Relying on No-Choice and Do not Repeat the Moves: Optimal, Efficient and Practical Algorithms for Assortment Optimization
Aadirupa Saha, Pierre Gaillard
International Conference on Learning Representations (ICLR), 2025 - Only Pay for What Is Uncertain: Variance-Adaptive Thompson Sampling.
Aadirupa Saha, Branislav Kveton
International Conference on Learning Representations (ICLR), 2024 - Eliciting User Preferences for Personalized Multi-Objective Decision Making through Comparative Feedback
Han Shao, Lee Cohen, Avrim Blum, Yishay Mansour, Aadirupa Saha, Mathew Walter
In Neural Information Processing Systems, NeurIPS 2023 - Federated Online and Bandit Convex Optimization [Arxiv Version]
Kumar Kshitij Patel, Lingxiao Wang, Aadirupa Saha, Nathan Srebro
In International Conference on Machine Learning, ICML 2023 - ANACONDA: Improved Dynamic Regret Algorithm for Adaptive Non-Stationary Dueling Bandits [Arxiv Version]
Thomas Kleine Buening, Aadirupa Saha
In International Conference on Artificial Intelligence and Statistics, AIStats 2023 - Versatile Dueling Bandits [Arxiv Version]
Aadirupa Saha, Pierre Gaillard
In International Conference on Machine Learning, ICML 2022 - Efficient and Optimal Algorithms for Contextual Dueling Bandits under Realizability [Arxiv Version]
Aadirupa Saha, Akshay Krishnamurthy
In Algorithmic Learning Theory, ALT 2022 - Optimal Algorithms for Stochastic Contextual Dueling Bandits
Aadirupa Saha
In Neural Information Processing Systems, NeurIPS 2021 - Dueling Convex Optimization
Aadirupa Saha, Tomer Koren, Yishay Mansour
In International Conference on Machine Learning, ICML 2021 - Optimal Regret Algorithm for Pseudo-1d Bandit Convex Optimization [Arxiv Version]
Aadirupa Saha, Nagarajan Natarajan, Praneeth Netrapalli, Prateek Jain
In International Conference on Machine Learning, ICML 2021 - From PAC to Instance-Optimal Sample Complexity in the Plackett-Luce Model [Arxiv Version]
Aadirupa Saha, Aditya Gopalan
In International Conference on Machine Learning, ICML 2020 - Combinatorial Bandits with Relative Feedback [Arxiv Version]
Aadirupa Saha, Aditya Gopalan
In Neural Information Processing Systems, NeurIPS 2019 - PAC Battling Bandits in the Plackett-Luce Model [Arxiv Version]
Aadirupa Saha, Aditya Gopalan
In Algorithmic Learning Theory, ALT 2019