FairTest: Discovering Unwarranted Associations in Data-Driven Applications

By Florian Tramèr, Vaggelis Atlidakis, Roxana Geambasu, Daniel J. Hsu, Jean-Pierre Hubaux, Mathias Humbert, Ari Juels, Huang Lin. 2015

191 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Safety Research, Evaluation & Benchmarks, AI Ethics, Fairness & Society

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Safety Research

In Evaluation & Benchmarks

In AI Ethics, Fairness & Society

Citations in the corpus

Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation (2017, neutral): “tected subgroup. Chouldechova and GâĂŹSell [11] trained two different outcome models then identified subgroups where the two models differed. Other papers work on a single outcome or risk score model [32, 35]. 2.2 Testing for Missing Features If the audit data is missing features used by the black-b”
AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Bias (2018, background): “FairTest (Tramèr et al., 2017), on the other hand, approaches the task of detecting biases in a dataset by checking for associations between predicted labels and protected attributes.”
Auditing Data Provenance in Text-Generation Models (2018, background): “improve the interpretability of the model by showing how features or training data points influence the model’s predictions. Other model-auditing research focuses on detecting bias and discrimination [30, 31]. We are not aware of any prior work that aims to audit the use of specific data sources to ”
Algorithmic Fairness (2020, neutral): “This dataset was studied, for example, in [144], [94] and [132], where age was the sensitive feature.”
Auditing Black-Box Models Using Transparent Model Distillation With Side Information (2017, background): “Other papers work on a single model, discovering subgroups that are miscalibrated when predicting outcomes [45], or exhibiting significant associations between protected features and black-box outputs [41].”
Themis: automatically testing software for discrimination (2018, methodology): “FairTest [31] uses manually written tests to measure four kinds of discrimination scores: the CV score and a related ratio, mutual”
Fairness testing: testing software for discrimination (2017, methodology): “FairTest, an implementation of the unwarranted associations framework [79] uses manually written tests to measure four kinds of discrimination scores: the CV score and a related ratio, mutual information, Pearson correlation, and a regression between the output and sensitive inputs.”
Ethical Considerations in AI-Based Recruitment (2019, methodology): “ FairTest [34]: FairTest is a toolkit to assess the features with the highest impact on a classifier (using an unwarranted association framework).”
Fairness-Aware Programming (2019, background): “The work on FairTest [23] provides a comprehensive framework for investigating fairness in data-driven pipelines.”
Machine Learning Testing: Survey, Landscapes and Horizons (2019, background): “[171] were the first to proposed the concept of ‘fairness bugs’.”
A Review on Fairness in Machine Learning (2022, unknown): “[198], where age was the sensitive feature.”
Interventional Fairness: Causal Database Repair for Algorithmic Fairness (2019, methodology): “Using this dataset, several prior efforts in algorithmic fairness have reported gender discrimination based on a strong statistical dependency between income and gender in favor of males [31, 48, 60].”
Trustworthy AI: From Principles to Practices (2021, background): “Criteria for AI systems have been studied to spot issues of unfairness by investigating the correlation among sensitive attributes, system outcomes, and the true label when applicable to well-designed diagnostic datasets [327].”
The Natural Auditor: How To Tell If Someone Used Your Words To Train Their Model (2018, background): “Another line of model-auditing research focuses on detecting bias and discrimination in models or data-driven applications [2], [35], [36].”