Which utilitarian ethical criteria does my model satisfy?
Consider the cautionary tale Automated Inference on Criminality using Face Images (WuZh16)
[…]we find some discriminating structural features for predicting criminality, such as lip curvature, eye inner corner distance, and the so-called nose-mouth angle. Above all, the most important discovery of this research is that criminal and non-criminal face images populate two quite distinctive manifolds. The variation among criminal faces is significantly greater than that of the non-criminal faces. The two manifolds consisting of criminal and non-criminal faces appear to be concentric, with the non-criminal manifold lying in the kernel with a smaller span, exhibiting a law of normality for faces of non-criminals. In other words, the faces of general law-biding public have a greater degree of resemblance compared with the faces of criminals, or criminals have a higher degree of dissimilarity in facial appearance than normal people.
There are so many problems with this. Which ones would you be happy with your local law enforcement authority taking home from this?
Maybe the in-progress textbook will have something to say? Solon Barocas, Moritz Hardt, Arvind Narayanan Fairness and machine learning.
Think pieces on fairness in models in practice
Fairness and trade offs
There are certain impossibility theorems around what you can simultaneously do here. However, that doesn't mean you can't fall short of the impossibility frontier on the side of unfairness (or straight up idiocy) if you don't work at it.
Chris Tucchio, at crunch conf makes some points about marginalist allocative/procedural fairness and net utility versus group rights.
If we choose to service Hyderabad with no disparities, we’ll run out of money and stop serving Hyderabad. The other NBFCs won’t.
Net result: Hyderabad is redlined by competitors and still gets no service.
Our choice: Keep the fraudsters out, utilitarianism over group rights.
He does a good job of explaining some impossibility theorems via examples, esp KlMR16. Note the interesting intersection of two types of classifications implicit in his model - uniformly reject, versus biassed accept/reject, subject to capital constraints. I need to revisit that and think some more.
Also I 'm interested in how this relates to the version of this problem with a time dimension…
Beauty contest problems
TODO: think about fairness problems that arise when the model is supposed to be rewarded on the basis of being a good bet for the future. Models that are supposed to predict credit risk are somewhat this - people in a poverty trap are bad credit risks, even if they got into the poverty trap because of lack of credit. A beauty contest problem is a model for this kind of situation, although there is a time-dimension also. There is presumable a game-theory equilibrium problem here. One imagines the Chinese restaurant process or something like it popping up, perhaps even the classic Pareto distribution or other Matthew Effect models.
- AgYu08: Charu C. Aggarwal, Philip S. Yu (2008) A General Survey of Privacy-Preserving Data Mining Models and Algorithms. In Privacy-Preserving Data Mining (pp. 11–52). Springer US DOI
- WuZh16: Xiaolin Wu, Xi Zhang (2016) Automated Inference on Criminality using Face Images. ArXiv:1611.04135 [Cs].
- KRPH17: Niki Kilbertus, Mateo Rojas-Carulla, Giambattista Parascandolo, Moritz Hardt, Dominik Janzing, Bernhard Schölkopf (2017) Avoiding Discrimination through Causal Reasoning. ArXiv:1706.02744 [Cs, Stat].
- BaSe16: Solon Barocas, Andrew D. Selbst (2016) Big Data’s Disparate Impact (SSRN Scholarly Paper No. ID 2477899). Rochester, NY: Social Science Research Network
- FFMS15: Michael Feldman, Sorelle A. Friedler, John Moeller, Carlos Scheidegger, Suresh Venkatasubramanian (2015) Certifying and Removing Disparate Impact. In Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 259–268). New York, NY, USA: ACM DOI
- Swee13: Latanya Sweeney (2013) Discrimination in Online Ad Delivery. Queue, 11(3), 10:10–10:29. DOI
- HaPS16: Moritz Hardt, Eric Price, Nati Srebro (2016) Equality of opportunity in supervised learning. In Advances in Neural Information Processing Systems (pp. 3315–3323).
- DHPR12: Cynthia Dwork, Moritz Hardt, Toniann Pitassi, Omer Reingold, Richard Zemel (2012) Fairness Through Awareness. In Proceedings of the 3rd Innovations in Theoretical Computer Science Conference (pp. 214–226). New York, NY, USA: ACM DOI
- Burr16: Jenna Burrell (2016) How the machine ‘thinks’: Understanding opacity in machine learning algorithms. Big Data & Society, 3(1), 2053951715622512. DOI
- KlMR16: Jon Kleinberg, Sendhil Mullainathan, Manish Raghavan (2016) Inherent Trade-Offs in the Fair Determination of Risk Scores.
- WPPA16: Scott Wisdom, Thomas Powers, James Pitton, Les Atlas (2016) Interpretable Recurrent Neural Networks Using Sequential Sparse Recovery. In Advances in Neural Information Processing Systems 29.
- ZWSP13: Rich Zemel, Yu Wu, Kevin Swersky, Toni Pitassi, Cynthia Dwork (2013) Learning Fair Representations. In Proceedings of the 30th International Conference on Machine Learning (ICML-13) (pp. 325–333).
- PRWK17: Geoff Pleiss, Manish Raghavan, Felix Wu, Jon Kleinberg, Kilian Q. Weinberger (2017) On Fairness and Calibration. In Advances In Neural Information Processing Systems.
- Mico17: Thomas Miconi (2017) The impossibility of “fairness”: a generalized impossibility result for decisions.