Perfectly Parallel Fairness Certification of Neural Networks
| Autor: | Valentin Wüstholz, Maria Christakis, Fuyuan Zhang, Caterina Urban |
|---|---|
| Přispěvatelé: | Département d'informatique - ENS Paris (DI-ENS), École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS), Analyse Statique par Interprétation Abstraite (ANTIQUE), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS-PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-Inria de Paris, Institut National de Recherche en Informatique et en Automatique (Inria), Max Planck Institute for Software Systems (MPI-SWS), ConsenSys Diligence, This work was supported by DFG grant 389792660 as part of TRR 248 (see https://perspicuous-computing.science)., Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Centre National de la Recherche Scientifique (CNRS)-Institut National de Recherche en Informatique et en Automatique (Inria)-École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Inria de Paris, Département d'informatique de l'École normale supérieure (DI-ENS), École normale supérieure - Paris (ENS Paris), Université Paris sciences et lettres (PSL)-Université Paris sciences et lettres (PSL)-Institut National de Recherche en Informatique et en Automatique (Inria)-Centre National de la Recherche Scientifique (CNRS)-École normale supérieure - Paris (ENS Paris), Caterina, Urban |
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
FOS: Computer and information sciences
Computer Science - Machine Learning Computer Science - Logic in Computer Science [INFO.INFO-LO] Computer Science [cs]/Logic in Computer Science [cs.LO] Fairness Neural Networks Computer science Distributed computing Abstract Interpretation 02 engineering and technology Certification Space (commercial competition) Machine Learning (cs.LG) Computer Science - Computers and Society Software Development (topology) [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG] [INFO.INFO-CY]Computer Science [cs]/Computers and Society [cs.CY] 020204 information systems Computers and Society (cs.CY) 0202 electrical engineering electronic engineering information engineering Safety Risk Reliability and Quality Computer Science - Programming Languages [INFO.INFO-PL]Computer Science [cs]/Programming Languages [cs.PL] Artificial neural network business.industry Static Analysis [INFO.INFO-LO]Computer Science [cs]/Logic in Computer Science [cs.LO] 020207 software engineering Static analysis Abstract interpretation [INFO.INFO-PL] Computer Science [cs]/Programming Languages [cs.PL] Logic in Computer Science (cs.LO) [INFO.INFO-CY] Computer Science [cs]/Computers and Society [cs.CY] Scalability business Programming Languages (cs.PL) |
| Zdroj: | Proceedings of the ACM on Programming Languages Proceedings of the ACM on Programming Languages, 2020, 4 (OOPSLA), pp.1-30. ⟨10.1145/3428253⟩ Proceedings of the ACM on Programming Languages, ACM, 2020, 4 (OOPSLA), pp.1-30. ⟨10.1145/3428253⟩ |
| ISSN: | 2475-1421 |
| DOI: | 10.1145/3428253⟩ |
| Popis: | International audience; Recently, there is growing concern that machine-learned software, which currently assists or even automates decision making, reproduces, and in the worst case reinforces, bias present in the training data. The development of tools and techniques for certifying fairness of this software or describing its biases is, therefore, critical. In this paper, we propose a perfectly parallel static analysis for certifying fairness of feed-forward neural networks used for classification of tabular data. When certification succeeds, our approach provides definite guarantees, otherwise, it describes and quantifies the biased input space regions. We design the analysis to be sound, in practice also exact, and configurable in terms of scalability and precision, thereby enabling pay-as-you-go certification. We implement our approach in an open-source tool called libra and demonstrate its effectiveness on neural networks trained on popular datasets. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|