Michiel A. Bakker
ABSTRACT
Prior work on fairness in machine learning has focused on settings where all the information needed about each individual is readily available. However, in many applications, further information may be acquired at a cost. For example, when assessing a customer's creditworthiness, a bank initially has access to a limited set of information but progressively improves the assessment by acquiring additional information before making a final decision. In such settings, we posit that a fair decision maker may want to ensure that decisions for all individuals are made with similar expected error rate, even if the features acquired for the individuals are different. We show that a set of carefully chosen confidence thresholds can not only effectively redistribute an information budget according to each individual's needs, but also serve to address individual and group fairness concerns simultaneously. Finally, using two public datasets, we confirm the effectiveness of our methods and investigate the limitations.
Supplemental Material
Available for Download
zip
aiespp093aux.zip (591.3 KB)
Includes the appendix to the main paper
- aono2017privacyAono, Y.; Hayashi, T.; Wang, L.; Moriai, S.; et al. 2017. Privacy-preserving deep learning via additively homomorphic encryption. IEEE Transactions on Information Forensics and Security 13(5): 1333--1345.Google Scholar
- Gummadi, Varshney, Weller, and Pentland]bakker2019dadiBakker, M. A.; Tu, D. P.; Valdés, H. R.; Gummadi, K. P.; Varshney, K. R.; Weller, A.; and Pentland, A. 2019. DADI: Dynamic Discovery of Fair Information with Adversarial Reinforcement Learning. arXiv preprint arXiv:1910.13983 .Google Scholar
- bechavod2017penalizingBechavod, Y.; and Ligett, K. 2017. Penalizing unfairness in binary classification. arXiv preprint arXiv:1707.00044 .Google Scholar
- berk2018fairnessBerk, R.; Heidari, H.; Jabbari, S.; Kearns, M.; and Roth, A. 2018. Fairness in criminal justice risk assessments: The state of the art. Sociological Methods & Research 0049124118782533.Google Scholar
- cai2020fairCai, W.; Gaebler, J.; Garg, N.; and Goel, S. 2020. Fair Allocation through Selective Information Acquisition. In Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society, 22--28.Google Scholar
- calmon2017optimizedCalmon, F.; Wei, D.; Vinzamuri, B.; Ramamurthy, K. N.; and Varshney, K. R. 2017. Optimized pre-processing for discrimination prevention. In Advances in Neural Information Processing Systems, 3992--4001.Google Scholar
- chen2018myChen, I.; Johansson, F. D.; and Sontag, D. 2018. Why Is My Classifier Discriminatory? In Advances in Neural Information Processing Systems, 3539.Google Scholar
- chouldechova2018caseChouldechova, A.; Benavides-Prado, D.; Fialko, O.; and Vaithianathan, R. 2018. A case study of algorithm-assisted decision making in child maltreatment hotline screening decisions. In Conference on Fairness, Accountability and Transparency, 134--148.Google Scholar
- chouldechova2018frontiersChouldechova, A.; and Roth, A. 2018. The frontiers of fairness in machine learning. arXiv preprint arXiv:1810.08810 .Google Scholar
- corbett2017algorithmicCorbett-Davies, S.; Pierson, E.; Feller, A.; Goel, S.; and Huq, A. 2017. Algorithmic decision making and the cost of fairness. In Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 797--806. ACM.Google Scholar
- davis2016privacyDavis, J.; and Osoba, O. 2016. Privacy Preservation in the Age of Big Data. Available at SSRN 2944731 .Google Scholar
- dwork2012fairnessDwork, C.; Hardt, M.; Pitassi, T.; Reingold, O.; and Zemel, R. 2012. Fairness through awareness. In Proceedings of the 3rd innovations in theoretical computer science conference, 214--226. ACM.Google Scholar
- eubanks2018automatingEubanks, V. 2018. Automating inequality: How high-tech tools profile, police, and punish the poor. St. Martin's Press.Google ScholarDigital Library
- feldman2015certifyingFeldman, M.; Friedler, S. A.; Moeller, J.; Scheidegger, C.; and Venkatasubramanian, S. 2015. Certifying and removing disparate impact. In proceedings of the 21th ACM SIGKDD international conference on knowledge discovery and data mining, 259--268.Google Scholar
- gao2011activeGao, T.; and Koller, D. 2011. Active classification based on value of classifier. In Advances in Neural Information Processing Systems.Google Scholar
- ez-Lobato]gong2019icebreakerGong, W.; Tschiatschek, S.; Turner, R.; Nowozin, S.; and Hernández-Lobato, J. M. 2019. Icebreaker: Element-wise Active Information Acquisition with Bayesian Deep Latent Gaussian Model. arXiv preprint arXiv:1908.04537 .Google Scholar
- hardt2016equalityHardt, M.; Price, E.; Srebro, N.; et al. 2016. Equality of opportunity in supervised learning. In Advances in neural information processing systems, 3315.Google Scholar
- rt-Johnson et al.(2018)Hébert-Johnson, Kim, Reingold, and Rothblum]hebert2018multicalibrationHébert-Johnson, Ú. ; Kim, M.; Reingold, O.; and Rothblum, G. 2018. Multicalibration: Calibration for the (computationally-identifiable) masses. In International Conference on Machine Learning, 1944--1953.Google Scholar
- sep-evidence-legalHo, H. L. 2015. The Legal Concept of Evidence. In Zalta, E. N., ed., The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University, winter 2015 edition.Google Scholar
- joseph2016fairnessJoseph, M.; Kearns, M.; Morgenstern, J. H.; and Roth, A. 2016. Fairness in learning: Classic and contextual bandits. In Advances in Neural Information Processing Systems, 325--333.Google Scholar
- kanani2008predictionKanani, P.; and Melville, P. 2008. Prediction-time active feature-value acquisition for cost-effective customer targeting. Advances In Neural Information Processing Systems (NIPS) .Google Scholar
- kaplan1968decisionKaplan, J. 1968. Decision Theory and the Factfinding Process. Stanford Law Review 1065--1092.Google Scholar
- kearns2017preventingKearns, M.; Neel, S.; Roth, A.; and Wu, Z. S. 2017. Preventing fairness gerrymandering: Auditing and learning for subgroup fairness. arXiv .Google Scholar
- krishnapuram2011costKrishnapuram, B.; Yu, S.; and Rao, R. B. 2011. Cost-sensitive Machine Learning. CRC Press.Google Scholar
- lewenberg2017knowingLewenberg, Y.; Bachrach, Y.; Paquet, U.; and Rosenschein, J. S. 2017. Knowing what to ask: A Bayesian active learning approach to the surveying problem. In Thirty-First AAAI Conference on Artificial Intelligence.Google Scholar
- lichman2013uciLichman, M.; et al. 2013. UCI machine learning repository.Google Scholar
- liu2008tefeLiu, L.-P.; Yu, Y.; Jiang, Y.; and Zhou, Z.-H. 2008. TEFE: A time-efficient approach to feature extraction. In Data Mining, 2008. ICDM'08. Eighth IEEE International Conference on. IEEE.Google Scholar
- ma2018eddiMa, C.; Tschiatschek, S.; Palla, K.; Hernandez-Lobato, J. M.; Nowozin, S.; and Zhang, C. 2019. EDDI: Efficient Dynamic Discovery of High-Value Information with Partial VAE. In International Conference on Machine Learning, 4234--4243.Google Scholar
- noriega2018activeNoriega-Campero, A.; Bakker, M.; Garcia-Bulle, B.; and Pentland, A. 2019. Active Fairness in Algorithmic Decision Making. Proceedings of AAAI / ACM Conference on Artificial Intelligence, Ethics, and Society .Google Scholar
- noriega2020algorithmicNoriega-Campero, A.; Garcia-Bulle, B.; Cantu, L. F.; Bakker, M. A.; Tejerina, L.; and Pentland, A. 2020. Algorithmic targeting of social policies: fairness, accuracy, and distributed governance. In Proceedings of the 2020 Conference on Fairness, Accountability, and Transparency, 241--251.Google Scholar
- platt1999probabilisticPlatt, J. C. 1999. Probabilistic outputs for support vector machines and comparisons to regularized likelihood methods. Advances in large margin classifiers 10(3): 61--74.Google Scholar
- pleiss2017fairnessPleiss, G.; Raghavan, M.; Wu, F.; Kleinberg, J.; and Weinberger, K. Q. 2017. On fairness and calibration. In Advances in Neural Information Processing Systems, 5680--5689.Google Scholar
- saar2007handlingSaar-Tsechansky, M.; and Provost, F. 2007. Handling missing values when applying classification models. Journal of machine learning research 8(Jul): 1623--1657.Google ScholarDigital Library
- kearns2019averageSharifi-Malvajerdi, S.; Kearns, M.; and Roth, A. 2019. Average Individual Fairness: Algorithms, Generalization and Experiments. In Advances in Neural Information Processing Systems, 8242--8251.Google Scholar
- shim2018jointShim, H.; Hwang, S. J.; and Yang, E. 2018. Joint active feature acquisition and classification with variable-size set encoding. In Advances in Neural Information Processing Systems, 1368--1378.Google Scholar
- speicher2018unifiedSpeicher, T.; Heidari, H.; Grgic-Hlaca, N.; Gummadi, K. P.; Singla, A.; Weller, A.; and Zafar, M. B. 2018. A unified approach to quantifying algorithmic unfairness: Measuring individual &group unfairness via inequality indices. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, 2239--2248.Google Scholar
- verma2018fairnessVerma, S.; and Rubin, J. 2018. Fairness definitions explained. In 2018 IEEE/ACM International Workshop on Software Fairness (FairWare), 1--7. IEEE.Google Scholar
- zafar2017fairnessZafar, M. B.; Valera, I.; Rogriguez, M. G.; and Gummadi, K. P. 2017. Fairness Constraints: Mechanisms for Fair Classification. In Artificial Intelligence and Statistics, 962--970.Google Scholar
- zhao2019secureZhao, C.; Zhao, S.; Zhao, M.; Chen, Z.; Gao, C.-Z.; Li, H.; and Tan, Y.-a. 2019. Secure multi-party computation: theory, practice and applications. Information Sciences 476: 357--372.Google ScholarCross Ref
- zhao2020individualZhao, S.; Ma, T.; and Ermon, S. 2020. Individual calibration with randomized forecasting. In International Conference on Machine Learning, 11387--11397. PMLR.Google Scholar
Index Terms
- Beyond Reasonable Doubt: Improving Fairness in Budget-Constrained Decision Making using Confidence Thresholds
Recommendations
On the apparent conflict between individual and group fairness
FAT* '20: Proceedings of the 2020 Conference on Fairness, Accountability, and TransparencyA distinction has been drawn in fair machine learning research between 'group' and 'individual' fairness measures. Many technical research papers assume that both are important, but conflicting, and propose ways to minimise the trade-offs between these ...
Relay-volunteered multi-rate cooperative MAC protocol for IEEE 802.11 WLANs
In IEEE 802.11, the rate of a station (STA) is dynamically determined by link adaptation. Low-rate STAs tend to hog more channel time than high-rate STAs due to fair characteristics of carrier sense multiple access/collision avoidance, leading to ...
Gerrymandering individual fairness
AbstractIndividual fairness requires that similar individuals are treated similarly. It is supposed to prevent the unfair treatment of individuals on the subgroup level and to overcome the problem that group fairness measures are susceptible to ...
Comments