Conference Papers

        1. FRAPPÉ: A Post-Processing Framework for Group Fairness Regularization Alexandru Țifrea, Preethi Lahoti, Ben Packer, Yoni Halpern, Ahmad Beirami, and Flavien Prost In International Conference on Machine Learning 2024 [Code]
        1. Can semi-supervised learning use all the data effectively? A lower bound perspective Alexandru Țifrea*, Gizem Yüce*, Amartya Sanyal, and Fanny Yang In Advances in Neural Information Processing Systems 2023 [Poster]
        2. Margin-based sampling in high dimensions: When being active is less efficient than staying passive Alexandru Țifrea*, Jacob Clarysse*, and Fanny Yang In International Conference on Machine Learning 2023 [Poster] [Video]
        1. Semi-supervised novelty detection using ensembles with regularized disagreement Alexandru Țifrea, Eric Petru Stavarache, and Fanny Yang In Conference on Uncertainty in Artificial Intelligence 2022 [Blog post] [Poster] [Code] [Video]
        1. Interpolation can hurt robust generalization even when there is no noise Konstantin Donhauser*, Alexandru Țifrea*, Michael Aerni, Reinhard Heckel, and Fanny Yang In Advances in Neural Information Processing Systems 2021 [Poster] [Video]
        1. Poincare Glove: Hyperbolic Word Embeddings Alexandru Țifrea, Gary Becigneul, and Octavian-Eugen Ganea In International Conference on Learning Representations 2019 [Blog post] [Poster] [Code]

          Preprints and Workshop Papers

                  1. Improving class and group imbalanced classification with uncertainty-based active learning Alexandru Țifrea*, John Hill*, and Fanny Yang In NeurIPS RealML Workshop 2023 [Poster]
                    1. Boosting worst-group accuracy without group annotations Vincent Bardenhagen*, Alexandru Țifrea*, and Fanny Yang In NeurIPS Workshop on Distribution Shifts 2021 [Poster]
                    2. Novel Disease Detection Using Ensembles with Regularized Disagreement Alexandru Țifrea, Eric Stavarache, and Fanny Yang In UNSURE Workshop at MICCAI 2021 [Code] [UNSURE@MICCAI2021] [Poster UNSURE]
                    3. Maximizing the robust margin provably overfits on noiseless data Konstantin Donhauser*, Alexandru Țifrea*, Michael Aerni, Reinhard Heckel, and Fanny Yang 2021 [Video] [AML@ICML21] [Poster AML] [OPPO@ICML21] [Poster OPPO]