Titre :

Justicia: A Stochastic SAT Approach to Formally Verify Fairness

Auteur(s) :

Ghosh, Bishwamittra [Auteur]
National University of Singapore [NUS]
Basu, Debabrota [Auteur]
Scool [Scool]
Chalmers University of Technology [Gothenburg, Sweden]
Meel, Kuldeep S. [Auteur]
National University of Singapore [NUS]

Titre de la manifestation scientifique :

AAAI Conference on Artificial Intelligence

Ville :

Virtual

Pays :

Canada

Date de début de la manifestation scientifique :

2021-02

Titre de l’ouvrage :

Proceedings of the AAAI Conference on Artificial Intelligence

Titre de la revue :

Proceedings of the AAAI Conference on Artificial Intelligence

Discipline(s) HAL :

Informatique [cs]/Apprentissage [cs.LG]
Informatique [cs]/Ordinateur et société [cs.CY]
Informatique [cs]/Logique en informatique [cs.LO]
Informatique [cs]/Intelligence artificielle [cs.AI]

Résumé en anglais : [en]

As a technology ML is oblivious to societal good or bad, and thus, the field of fair machine learning has stepped up to propose multiple mathematical definitions, algorithms, and systems to ensure different notions of ...
Lire la suite >As a technology ML is oblivious to societal good or bad, and thus, the field of fair machine learning has stepped up to propose multiple mathematical definitions, algorithms, and systems to ensure different notions of fairness in ML applications. Given the multitude of propositions, it has become imperative to formally verify the fairness metrics satisfied by different algorithms on different datasets. In this paper, we propose a stochastic satisfiability (SSAT) framework, Justicia, that formally verifies different fairness measures of supervised learning algorithms with respect to the underlying data distribution. We instantiate Justicia on multiple classification and bias mitigation algorithms, and datasets to verify different fairness metrics, such as disparate impact, statistical parity, and equalized odds. Justicia is scalable, accurate, and operates on non-Boolean and compound sensitive attributes unlike existing distribution-based verifiers, such as FairSquare and VeriFair. Being distribution-based by design, Justicia is more robust than the verifiers, such as AIF360, that operate on specific test samples. We also theoretically bound the finite-sample error of the verified fairness measure.Lire moins >

Langue :

Anglais

Comité de lecture :

Oui

Audience :

Internationale

Vulgarisation :

Non

Références liée(s) :

https://github.com/meelgroup/justicia

Collections :

• Centre de Recherche en Informatique, Signal et Automatique de Lille (CRIStAL) - UMR 9189

Source :

Harvested from HAL

Date de dépôt :

2021-11-26T02:00:54Z