Fairness Testing: Testing Software for Discrimination
Sainyam Galhotra, Yuriy Brun, Alexandra Meliou

TL;DR
This paper introduces Themis, a testing method to measure software discrimination, revealing that many systems exhibit significant bias and existing bias removal techniques often fail, emphasizing the importance of fairness testing.
Contribution
The paper presents Themis, an automated, input-schema-based testing tool for measuring software discrimination without requiring an oracle, and evaluates its effectiveness across multiple systems.
Findings
Themis effectively discovers software discrimination.
State-of-the-art bias removal techniques often fail, sometimes discriminating against 98%.
Themis produces efficient test suites, especially on systems with higher discrimination.
Abstract
This paper defines software fairness and discrimination and develops a testing-based method for measuring if and how much software discriminates, focusing on causality in discriminatory behavior. Evidence of software discrimination has been found in modern software systems that recommend criminal sentences, grant access to financial products, and determine who is allowed to participate in promotions. Our approach, Themis, generates efficient test suites to measure discrimination. Given a schema describing valid system inputs, Themis generates discrimination tests automatically and does not require an oracle. We evaluate Themis on 20 software systems, 12 of which come from prior work with explicit focus on avoiding discrimination. We find that (1) Themis is effective at discovering software discrimination, (2) state-of-the-art techniques for removing discrimination from algorithms fail…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
