André F. Cruz

Max Planck Institute for Intelligent Systems

profile/prof_pic.jpg

Office: N1.019

Max-Planck-Ring 4

Tübingen, Germany

Hi!

I’m a PhD student at the MPI for Intelligent Systems (MPI-IS), in Germany. I’m part of the Social Foundations of Computation Department, where I’m fortunate to be advised by Moritz Hardt.

My current research focus is on the fair and responsible use of Machine Learning, and the feedback loops between deployed ML systems and society at large.

Before joining MPI-IS, I was a Research Scientist on the Responsible AI group at Feedzai, a fintech startup :unicorn: where I researched and productized fair ML in real-world settings.

news

Sep. 2024 Our paper on “Evaluating language models as risk scores” was accepted to NeurIPS 2024 :bookmark:
Jun. 2024 We open-sourced the folktexts package: a benchmark for LLMs on tabular data tasks :trophy:
Jan. 2024 Our paper “Unprocessing Seven Years of Algorithmic Fairness” was accepted as an oral at ICLR 2024 (notable top 5%) :tada:

selected publications

  1. Evaluating language models as risk scores
    André F. Cruz, Moritz Hardt, and Celestine Mendler-Dünner
    In The Thirty-eight Conference on Neural Information Processing Systems Datasets and Benchmarks Track (NeurIPS), 2024
  2. Unprocessing Seven Years of Algorithmic Fairness
    André F Cruz, and Moritz Hardt
    In The Twelfth International Conference on Learning Representations (ICLR), 2024 (Oral Presentation)
  3. FairGBM: Gradient Boosting with Fairness Constraints
    André F Cruz, Catarina Belém, Sérgio Jesus, and 3 more authors
    In The Eleventh International Conference on Learning Representations (ICLR), 2023
  4. Promoting Fairness through Hyperparameter Optimization
    André F Cruz, Pedro Saleiro, Catarina Belém, and 2 more authors
    In 2021 IEEE International Conference on Data Mining (ICDM), 2021
  5. TimeSHAP: Explaining Recurrent Models through Sequence Perturbations
    João Bento, Pedro Saleiro, André F Cruz, and 2 more authors
    In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining, 2021