Can machines discriminate? And how to prevent that

Abstract:

Discrimination-aware data mining/ machine learning is an emerging discipline at an intersection of computer science, law and social sciences. Law defines no-discrimination constraints, machine learning incorporates those constraints into algorithms for learning models from data. The goal is to prevent discrimination when decisions in real life are made based on model outputs. As an example consider automated CV screening for new openings at job portals. This talk will give an overview of current research trends in discrimination-aware machine learning.

Bio:

Indrė Žliobaitė is a researcher at Aalto University, HIIT, and University of Helsinki. Her main research interests include predictive modeling with streaming data, spatio-temporal data, and predictive modeling with no-discrimination constraints. For more information see http://www.zliobaite.com/