Algorithmic fairness: why it’s hard and why it’s interesting

In only a few years, algorithmic fairness has grown from a niche topic to a major component of computer vision, machine learning and artificial intelligence research and practice. As a field, we have had some embarrassing mistakes, yet our understanding of the core issues, potential impacts, and mitigation approaches has grown. This tutorial presents a range of recent findings, discussions, questions, and partial answers in the space of algorithmic fairness in recent years. While this tutorial will not attempt a comprehensive overview of this rich area, we aim to provide the participants with some tools and insights and to explore the connections between algorithmic fairness and a broad range of ongoing research efforts in the field. We will tackle some of the hard questions that you may have about algorithmic fairness, and hopefully address some misconceptions that have become pervasive.

The tutorial is designed to be accessible to a broad audience of computer vision and artificial intelligence researchers and practitioners.

Agenda and logistics:

The tutorial will take place on Sunday, June 19th in Room 06-09 and on zoom (see the CVPR Virtual Site for a link).

We will start at 1:30pm and will run until about 5pm, with a 30-minute break.

The tutorial will be structured around 7 narratives of algorithmic fairness. For a taster, "Narrative 2: Once you define the metric, algorithmic fairness is straightforward." We will discuss each narrative in turn, first providing some context and then diving into the arguments that can be made for and against. Our plan is to be somewhat controversial and present some "devil's advocate" arguments -- while also tackling some of the more challenging questions in this space.

We will aim to leave plenty of time at the end for discussion.