July 2nd, 2.15 pm - 3 pm, Tom Sühr will give the talk: "Fairness preserving interfaces for rankings" at the HCC Lab.

Tom Sühr is a Masters student of Information Systems Management at TU Berlin. He is currently working on research projects with Harvard University, Pompeu Fabra University and the Max Planck Institute for Software Systems. During his undergraduate studies he was a research intern at the Max Planck Institute for Software Systems advised by Krishna Gummadi and Asia Biega as well as a student researcher at the chair of complex distributed systems (CIT) advised by Meike Zehlike and Carlos Castillo. He is broadly interested in algorithmic fairness, fair systems and economic implications of algorithmic decision making.

Abstract

In our everyday life we make decisions based on rankings. We decide where we work, what we consume and who we hire based on rankings presented by platforms like Google, Amazon, LinkedIn and Netflix. Past research has shown that those systems have a tendency to discrimination based on protected features like gender or ethnicity. In order to make those systems fair, projects tackled the problems through so called pre-processing, in-processing and post-processing methods. However in many cases, the final decision is up to the user. If the user decision is biased, this might end in unfair results for the ranked items/people. It is important to ask, how a bias in user decision making impacts a fair algorithm and how to build user interfaces that preserve the fairness of a fair algorithm. In this talk, we will explore how we want to measure bias in human decision making, as well as possible pitfalls in the current evaluation of fair algorithms.

This is joint work in progress with Prof. Himabindu Lakkaraju (Harvard University), Prof. Walter Lasecki (University of Michigan) and Sophie Hilgard (Harvard University).

Date: July 2nd, 2020
Location: If you want to join the presentation, please contact Dr.-Ing. Christoph Kinkeldey.