Abstract
Artificial Intelligence and Machine Learning are becoming increasingly present in several aspects of human life, especially, those dealing with decision making. Many of these algorithmic decisions are taken without human supervision and through decision making processes that are not transparent. This raises concerns regarding the potential bias of these processes towards certain groups of society, which may entail unfair results and, possibly, violations of human rights. Dealing with such biased models is one of the major concerns to maintain the public trust.
In this paper, we address the question of process or procedural fairness. More precisely, we consider the problem of making classifiers fairer by reducing their dependence on sensitive features while increasing (or, at least, maintaining) their accuracy. To achieve both, we draw inspiration from “dropout” techniques in neural based approaches, and propose a framework that relies on “feature drop-out” to tackle process fairness. We make use of “LIME Explanations” to assess a classifier’s fairness and to determine the sensitive features to remove. This produces a pool of classifiers (through feature dropout) whose ensemble is shown empirically to be less dependent on sensitive features, and with improved or no impact on accuracy.
This research was partially supported by TAILOR, a project funded by EU Horizon 2020 research and innovation programme under GA No. 952215, and the Inria Project Lab “Hybrid Approaches for Interpretable AI” (HyAIAI).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
- 2.
General Data Protection Regulation (GDPR): https://gdpr-info.eu/.
- 3.
Terms unfairness and bias are used interchangeably.
- 4.
- 5.
- 6.
LableEncoder Class is given in the sklearn preprocessing library
https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.LabelEncoder.html.
- 7.
- 8.
- 9.
Here we focus on binary classifiers that output the probability for each class label.
- 10.
- 11.
In this study we focused on the top 10 features. However this parameter can be set by the user and changed according to his use case.
- 12.
Adult Dataset: http://archive.ics.uci.edu/ml/datasets/Adult.
- 13.
- 14.
- 15.
We performed the t-test.
- 16.
- 17.
It depicts if a person gave aphone number. Due to privacy reasons, the number may not be given. Thus it should not be considered important.
- 18.
Interestingly, there is an accuracy increase when that variable is dropped. However, the current implementation of LimeOut does not take action in these cases.
References
Binns, R.: On the apparent conflict between individual and group fairness. In: Conference on Fairness, Accountability, and Transparency (FAT20), pp. 514–524 (2020)
Chouldechova, A.: Fair prediction with disparate impact: a study of bias in recidivism prediction instruments. Big Data 5(2), 153–163 (2017)
Dressel, J., Farid, H.: The accuracy, fairness, and limits of predicting recidivism. Sci. Adv. 4(eaao5580), 1 (2018)
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.S.: Fairness through awareness. In: Goldwasser, S. (ed.) Innovations in Theoretical Computer Science 2012, Cambridge, MA, USA, January 8–10, 2012, pp. 214–226. ACM (2012)
Gal, Y., Ghahramani, Z.: Dropout as a Bayesian approximation: representing model uncertainty in deep learning. In: Balcan, M., Weinberger, K.Q. (eds.) International Conference on Machine Learning, ICML16. JMLR Workshop and Conference Proceedings, vol. 48, pp. 1050–1059 (2016)
Gal, Y., Ghahramani, Z.: A theoretically grounded application of dropout in recurrent neural networks. In: Lee, D.D., Sugiyama, M., von Luxburg, U., Guyon, I., Garnett, R. (eds.) Neural Information Processing Systems (NIPS16), pp. 1019–1027 (2016)
Garreau, D., von Luxburg, U.: Explaining the explainer: a first theoretical analysis of LIME. CoRR abs/2001.03447 (2020)
Grgic-Hlaca, N., Redmiles, E.M., Gummadi, K.P., Weller, A.: Human perceptions of fairness in algorithmic decision making: a case study of criminal risk prediction. In: World Wide Web (WWW18), pp. 903–912 (2018)
Grgic-Hlaca, N., Zafar, M.B., Gummadi, K.P., Weller, A.: The case for process fairness in learning: feature selection for fair decision making. In: NIPS Symposium on Machine Learning and the Law, vol. 1, p. 2 (2016)
Grgić-Hlača, N., Zafar, M.B., Gummadi, K.P., Weller, A.: Beyond distributive fairness in algorithmic decision making: feature selection for procedurally fair learning. In: Proceedings of the Conference on Artificial Intelligence (AAAI18), pp. 51–60 (2018)
Guegan, D., Addo, P.M., Hassani, B.: Credit risk analysis using machine and deep learning models. Risks 6(2), 38 (2018)
Iskandar, B.: Terrorism detection based on sentiment analysis using machine learning. J. Eng. Appl. Sci. 12(3), 691–698 (2017)
Kearns, M., Neel, S., Roth, A., Wu, Z.S.: Preventing fairness gerrymandering: auditing and learning for subgroup fairness. In: International Conference on Machine Learning (ICML18), pp. 2564–2572 (2018)
Laugel, T., Renard, X., Lesot, M.J., Marsala, C., Detyniecki, M.: Defining locality for surrogates in post-hoc interpretablity. arXiv preprint arXiv:1806.07498 (2018)
van der Linden, I., Haned, H., Kanoulas, E.: Global aggregations of local explanations for black box models. arXiv abs/1907.03039 (2019)
Lundberg, S.M., Lee, S.: A unified approach to interpreting model predictions. In: Conference on Neural Information Processing Systems (NIPS17), pp. 4765–4774 (2017)
Maes, S., Tuyls, K., Vanschoenwinkel, B., Manderick, B.: Credit card fraud detection using Bayesian and neural networks. In: NAISO Congress on Neuro Fuzzy Technologies, pp. 261–270 (2002)
Ribeiro, M.T., Singh, S., Guestrin, C.: “Why should i trust you?”: explaining the predictions of any classifier. In: International Conference on Knowledge Discovery and Data Mining (SIGKDD16), pp. 1135–1144 (2016)
Ribeiro, M.T., Singh, S., Guestrin, C.: Anchors: high-precision model-agnostic explanations. In: AAAI Conference on Artificial Intelligence, AAAI18, pp. 1527–1535 (2018)
Roh, Y., Heo, G., Whang, S.E.: A survey on data collection for machine learning: a big data - AI integration perspective. arXiv abs/1811.03402 (2018)
Speicher, T., et al.: A unified approach to quantifying algorithmic unfairness: measuring individual & group unfairness via inequality indices. In: International Conference on Knowledge Discovery & Data Mining (SIGKDD18), pp. 2239–2248 (2018)
Zafar, M.B., Valera, I., Gomez Rodriguez, M., Gummadi, K.P.: Fairness beyond disparate treatment & disparate impact: learning classification without disparate mistreatment. In: World Wide Web (WWW17), pp. 1171–1180 (2017)
Zafar, M.B., Valera, I., Rodriguez, M.G., Gummadi, K.P.: Fairness constraints: mechanisms for fair classification. In: Artificial Intelligence and Statistics (AISTATS17), pp. 962–970 (2017)
Zemel, R., Wu, Y., Swersky, K., Pitassi, T., Dwork, C.: Learning fair representations. In: International Conference on Machine Learning (ICML13), pp. 325–333 (2013)
Zhang, Z., Neill, D.B.: Identifying significant predictive bias in classifiers. CoRR abs/1611.08292 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Bhargava, V., Couceiro, M., Napoli, A. (2020). LimeOut: An Ensemble Approach to Improve Process Fairness. In: Koprinska, I., et al. ECML PKDD 2020 Workshops. ECML PKDD 2020. Communications in Computer and Information Science, vol 1323. Springer, Cham. https://doi.org/10.1007/978-3-030-65965-3_32
Download citation
DOI: https://doi.org/10.1007/978-3-030-65965-3_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-65964-6
Online ISBN: 978-3-030-65965-3
eBook Packages: Computer ScienceComputer Science (R0)