Machine Learning and Causal Inference for Reliable Decision-Making in High-Stakes Settings

Project Description

Machine Learning (ML) systems are becoming instrumental across many domains, with applications spanning areas like criminal justice, social welfare, financial fraud detection, and public health. While these systems offer potential benefits to institutional decision-making processes, such as improved efficiency and reliability, they still face the challenge of aligning intricate and nuanced policy objectives with the precise formalization requirements necessitated by the ML model. In this project, we aim to enhance the reliability of statistical targeting procedures used for resource allocation problems by incorporating notions from algorithmic fairness, specifically multi-calibration and multi-accuracy.

Publications

• Fischer Abaigar, Unai, Christoph Kern, Noam Barda, and Frauke Kreuter. “Bridging the Gap: Towards an Expanded Toolkit for ML-Supported Decision-Making in the Public Sector.” arXiv, October 29, 2023. https://doi.org/10.48550/arXiv.2310.19091.