Fair-doctor: Detecting and mitigating unfairness in neural networks

Adhikari, Rittika

Permalink

https://hdl.handle.net/2142/113921 Copy

Description

Title

Fair-doctor: Detecting and mitigating unfairness in neural networks

Author(s)

Adhikari, Rittika

Issue Date

2021-12-09

Director of Research (if dissertation) or Advisor (if thesis)

Koyejo, Oluwasanmi

Department of Study

Computer Science

Discipline

Computer Science

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• machine learning
• fairness
• interpretability

Abstract

"Important decisions are increasingly based directly on predictions from classifiers; for example, machine learning models are now being used to facilitate autonomous vehicles, predict stock market trends, diagnose illnesses, and so much more. However, users fundamentally understand very little about how these black box classifiers come to make decisions, and whether these predictions are unbiased. With the more prevalent adoption of these systems, it is crucial that we must be able to both explain and understand what concepts our models utilize to make predictions to ensure that we are building unbiased, interpretable models. To facilitate this, we propose Fair-Doctor, a pipeline which diagnoses unfairness, treats it, and follows up to ensure that the algorithmic bias has been mitigated. We utilize TCAV (Testing With Concept Activation Vectors), a state-of-the-art interpretability tool, to diagnose unfairness. We also introduce a novel adversarial fairness loss function, which works to remove the specified unfairness in the model. We evaluate this architecture on a simple CNN trained on CelebA to predict how ""young"" a person looks. Our results demonstrate that we are able to successfully reduce the bias in this model."

Graduation Semester

2021-12

Type of Resource

Thesis

Permalink

http://hdl.handle.net/2142/113921

Copyright and License Information

Copyright 2021 Rittika Adhikari

Owning Collections