EURO 2024 Copenhagen
Abstract Submission

EURO-Online login

4017. Making Black-Boxes Explainable and Fair Through Mathematical Optimization

Invited abstract in session MB-27: On Mathematical Optimization for Explainable and Fair Machine Learning, stream Mathematical Optimization for XAI.

Monday, 10:30-12:00
Room: 047 (building: 208)

Authors (first author is the speaker)

1. Thomas Halskov
Copenhagen Business School
2. Emilio CARRIZOSA
IMUS - Instituto de Matemáticas de la Universidad de Sevilla
3. Dolores Romero Morales
Copenhagen Business School

Abstract

Machine Learning has increasingly been used for algorithmic decision-making, often through what are considered "black-boxes", i.e., models which do not provide an explanation for their outputs. In this talk, we consider the task of explaining the output and decisions from a given "black-box" to provide an interpretable explanation. Additionally, we tackle fairness concerns that might arise from using "black-box" models, where discrimination can be present. Our approach aims to give a local explanation for an output, while taking into account global fairness and accuracy.

Keywords

Status: accepted


Back to the list of papers