2020

Ethical Adversaries: Towards Mitigating Unfairness with Adversarial Machine Learning

Delobelle, Pieter, Temple, Paul, Perrouin, Gilles et al.

Understand

Machine learning is being integrated into a growing number of critical systems with far-reaching impacts on society.

Unexpected behaviour and unfair decision processes are coming under increasing scrutiny due to this widespread use and its theoretical considerations.
Individuals, as well as organisations, notice, test, and criticize unfair results to hold model designers and deployers accountable.
We offer a framework that assists these groups in mitigating unfair representations stemming from the training datasets.

Built on

In: Kdd. vol. 96, pp. 202–207 (1996)
Kohavi, R.: Scaling up the accuracy of naive-bayes classifiers: A decision-tree hybrid · 1996
Earlier work this paper cites.
Data Min. Knowl. Discov. 21(2), 277–292 (2010)
Calders, T., Verwer, S.: Three naive Bayes approaches for discrimination-free classification · 2010
Earlier work this paper cites.
In: Proceedings of the 29th International Conference on Machine Learning. pp. 1467–1474. ICML’12 (2012)
Biggio, B., Nelson, B., Laskov, P.: Poisoning attacks against support vector machines · 2012
Earlier work this paper cites.
In: 3rd Innovations in Theoretical Computer Science Conference. pp. 214–226. ACM (2012)
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., Zemel, R.: Fairness Through Awareness · 2012
Earlier work this paper cites.
IEEE Trans. Knowl. Data Eng. 25(7), 1445–1459 (2013), https://doi.org/10.1109/TKDE.2012.72
Hajian, S., Domingo-Ferrer, J.: A methodology for direct and indirect discrimination prevention in data mining · 2012
Earlier work this paper cites.
In: 2013 International Conference on Biometrics (ICB). pp. 1–7 (2013a)
Biggio, B., Didaci, L., Fumera, G., Roli, F.: Poisoning attacks to compromise face templates · 2013
Earlier work this paper cites.
In: International Conference on Machine Learning. pp. 325–333 (2013)
Zemel, R., Wu, Y., Swersky, K., Pitassi, T., Dwork, C.: Learning fair representations · 2013
Earlier work this paper cites.
In: NIPS, pp. 2672–2680. Curran Associates (2014)
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., Bengio, Y.: Generative Adversarial Nets · 2014
Earlier work this paper cites.
arXiv:1511.05897 (2015)
Original
Edwards, H., Storkey, A.: Censoring Representations with an Adversary · 2015
Earlier work this paper cites.
In: 21th ACM SIGKDD International Conference. pp. 259–268 (2015)
Feldman, M., Friedler, S.A., Moeller, J., Scheidegger, C., Venkatasubramanian, S.: Certifying and removing disparate impact · 2015
Earlier work this paper cites.
ProPublica (2016)
Angwin, J., Larson, J.: Machine bias: There’s software used across the country to predict future criminals. and it’s biased against blacks · 2016
Earlier work this paper cites.
Dieterich, W., Mendoza, C., Brennan, T.: COMPAS Risk Scales: Demonstrating Accuracy Equity and Predictive Parity (2016)
2016
Earlier work this paper cites.

Similar

The Journal of Machine Learning Research 17(1), 2096–2030 (2016)
Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., Lempitsky, V.: Domain-adversarial training of neural networks · 2016
Cited alongside, same era.
In: NIPS, pp. 3315–3323. Curran Associates (2016)
Hardt, M., Price, E., ecprice, Srebro, N.: Equality of Opportunity in Supervised Learning · 2016
Cited alongside, same era.
In: IEEE European Symposium on Security and Privacy. pp. 372–387 (2016)
Papernot, N., McDaniel, P., Jha, S., Fredrikson, M., Celik, Z.B., Swami, A.: The limitations of deep learning in adversarial settings · 2016
Cited alongside, same era.
arXiv:1703.00056 (2017)
Original
Chouldechova, A.: Fair prediction with disparate impact: A study of bias in recidivism prediction instruments · 2017
Cited alongside, same era.
arXiv:1706.09847 (2017)
Original
Ensign, D., Friedler, S.A., Neville, S., Scheidegger, C., Venkatasubramanian, S.: Runaway Feedback Loops in Predictive Policing · 2017
Cited alongside, same era.
In: Asian Conference on Computer and Communications Security. pp. 506–519. ACM (2017)
Papernot, N., McDaniel, P., Goodfellow, I., Jha, S., Celik, Z.B., Swami, A.: Practical black-box attacks against machine learning · 2017
Cited alongside, same era.
Pattern Recognition Journal 84, 317–331 (2018)
Biggio, B., Roli, F.: Wild patterns: Ten years after the rise of adversarial machine learning · 2018
Cited alongside, same era.
arXiv:1808.00023 (2018)
Original
Corbett-Davies, S., Goel, S.: The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning · 2018
Cited alongside, same era.
In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing. pp. 11–21. ACL, Brussels, Belgium (10)
Elazar, Y., Goldberg, Y.: Adversarial removal of demographic attributes from text data · 2018
Cited alongside, same era.
arXiv abs/1802.06309 (2018)
Original
Madras, D., Creager, E., Pitassi, T., Zemel, R.S.: Learning Adversarially Fair and Transferable Representations · 2018
Cited alongside, same era.
arXiv:1811.11293 (2018)
Original
Overdorf, R., Kulynych, B., Balsa, E., Troncoso, C., Gürses, S.: Questioning the assumptions behind fairness solutions · 2018
Cited alongside, same era.
In: ECML/PKDD. pp. 387–402 (2013b)
Biggio, B., Corona, I., Maiorca, D., Nelson, B., Šrndić, N., Laskov, P., Giacinto, G., Roli, F.: Evasion attacks against machine learning at test time
Cited in the paper.

Then

In: Storkey, A., Perez-Cruz, F. (eds.) 21st International Conference on Artificial Intelligence and Statistics. Proceedings of Machine Learning Research, vol. 84, pp. 96–105. PMLR, Playa Blanca, Lanzarote, Canary Islands (2018)
Pierson, E., Corbett-Davies, S., Goel, S.: Fast threshold tests for detecting discrimination · 2018
Later among the works it cites.
In: IEEE 5th International Conference on Data Science and Advanced Analytics. pp. 189–198 (2018)
Raff, E., Sylvester, J.: Gradient Reversal against Discrimination: A Fair Neural Network Learning Approach · 2018
Later among the works it cites.
In: Conference on AI, Ethics, and Society. pp. 243–250 (2018)
Raff, E., Sylvester, J., Mills, S.: Fair forests: Regularized tree induction to minimize model bias · 2018
Later among the works it cites.
In: Proceedings of Conference on AI, Ethics, and Society. pp. 335–340. ACM Press (2018)
Zhang, B.H., Lemoine, B., Mitchell, M.: Mitigating Unwanted Biases with Adversarial Learning · 2018
Later among the works it cites.
In: AAAI Conference on Artificial Intelligence (2019)
Adel, T., Valera, I., Ghahramani, Z., Weller, A.: One-Network Adversarial Fairness · 2019
Later among the works it cites.
fairmlbook.org (2019)
Barocas, S., Hardt, M., Narayanan, A.: Fairness and Machine Learning · 2019
Later among the works it cites.
In: 28th USENIX Security Symposium. pp. 321–338 (2019)
Demontis, A., Melis, M., Pintor, M., Jagielski, M., Biggio, B., Oprea, A., Nita-Rotaru, C., Roli, F.: Why do adversarial attacks transfer? Explaining transferability of evasion and poisoning attacks · 2019
Later among the works it cites.
IBM Journal of Res. and Dev. p. 12 (2019)
Sattigeri, P., Hoffman, S.C., Chenthamarakshan, V., Varshney, K.R.: Fairness GAN: Generating Datasets With Fairness Properties Using a Generative Adversarial Network · 2019
Later among the works it cites.
arXiv:2005.14050 [cs] (May 2020)
Original
Blodgett, S.L., Barocas, S., Daumé III, H., Wallach, H.: Language (Technology) is Power: A Critical Survey of ”Bias” in NLP · 2020
Closest in time.
Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency pp. 177–188 (Jan 2020)
Kulynych, B., Overdorf, R., Troncoso, C., Gürses, S.: POTs: Protective Optimization Technologies · 2020
Closest in time.
arXiv preprint arXiv:2004.07401 (2020)
Original
Solans, D., Biggio, B., Castillo, C.: Poisoning attacks on algorithmic fairness · 2020
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…