Konstantina Bairaktari

konstantina@cs.au.dk
she/her

I am a Postdoctoral Researcher at the Algorithms, Data Structures and Foundations of Machine Learning Group at Aarhus University, hosted by Kasper Green Larsen. I completed my PhD at Northeastern University, where I was advised by Huy Le Nguyen and Jonathan Ullman. My research interests are in theoretical approaches to fairness in machine learning and data privacy.

Publications

Testable and Actionable Calibration for Full Swap Regret

Konstantina Bairaktari, Lunjia Hu, Huy Le Nguyen and Jonathan Ullman

Preprint

Sample-efficient Multiclass Calibration under L_p Error

Konstantina Bairaktari and Huy Le Nguyen

Preprint

paper

Kandinsky Conformal Prediction: Beyond Class- and Covariate-Conditional Coverage

Konstantina Bairaktari, Jiayun Wu and Zhiwei Steven Wu

International Conference on Machine Learning (ICML 2025)

paper

Privacy in Metalearning and Multitask Learning: Modeling and Separations

Maryam Aliakbarpour, Konstantina Bairaktari, Adam Smith, Marika Swanberg and Jonathan Ullman

International Conference on Artificial Intelligence and Statistics (AISTATS 2025)

paper

Metalearning with Very Few Samples Per Task

Maryam Aliakbarpour, Konstantina Bairaktari, Gavin Brown, Adam Smith, Nathan Srebro and Jonathan Ullman

Conference on Learning Theory (COLT 2024)

paper

Multitask Learning via Shared Features: Algorithms and Hardness

Konstantina Bairaktari, Guy Blanc, Li-Yang Tan, Jonathan Ullman and Lydia Zakynthinou

Conference on Learning Theory (COLT 2023)

paper

Fair and Useful Cohort Selection

Konstantina Bairaktari, Paul Langton, Huy Le Nguyen, Niklas Smedemark-Margulies and Jonathan Ullman

Transactions on Machine Learning Research (TMLR 2023)

paper