Fairness in Machine Learning

Course Description

We will present frameworks for auditing and mitigating biased decision making by machine learning models. We will discuss the accuracy fairness trade-off. Lastly, we will introduce students a practical tool to define and measure the fairness of machine learning algorithms using our EthicML toolbox (https://github.com/predictive-analytics-lab/EthicML).

Course tools

  • Python

Prerequisites

  • Supervised machine learning

Level of complexity of course

Intermediate

Lecturer

Dr. Novi Quadrianto

Since May 2019 Novi Quadrianto is a Reader in Machine Learning at the University of Sussex, UK. Prior to Sussex, she was a Newton International Fellow of the Royal Society and the British Academy at the Machine Learning Group in the Department of Engineering, University of Cambridge. Novi received my Ph.D. in Computer Science from the Australian National University, Canberra, Australia in July 2012. In 2019, she was awarded a European Research Council ERC Starting Grant for a project on developing Bayesian models and algorithms for fairness and transparency (BayesianGDPR).

Fields of interests: Machine Learning, Algorithmic Fairness, and Transparency, Causality.

Contacts[email protected]