7 ms·
Delayed Impact of Fair Machine Learning
- said 9y agoTo authors Max Simchowitz and Moritz Hardt: It’s time we wield your fair weapons against your own tribe: Jews. Jews are dramatically over-represented in positions of power, influence, wealth, and information gate-keeping (e.g. journalism, entertainment media, the Supreme Court, Congress, academia, the Federal Reserve, the top 1%). Let’s give everyone institutional affirmative action at your expense! Yes, Jews who are struggling will be the ones the hardest hit. But that’s okay! What matters is “demographic parity” drawn along lines you usually pretend are too blurry to define.
- arbie 9y agoYour incendiary remarks are unrelated to the article and are not welcome on HN.
- said 9y agoThis article calls for structural affirmative action to benefit blacks at the expense of non-blacks. Wherever this weapon is yielded, be it Harvard admissions or tech leadership within Google, a pattern emerges regarding which groups are most and least affected. It’s in poor taste to propose more of these policies while being a member of an unaffected group.
- yorwba 9y agoThe article is actually about benefiting credit-worthy people with a bad credit score by giving them loans they can pay back, and they show that attempting to enforce certain fairness criteria is actually counter to that.
- dang 9y agoWe've banned this account for repeatedly violating the HN guidelines and ignoring our requests to stop. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html
- joe_the_user 9y ago"Machine learning systems trained to minimize prediction error may often exhibit discriminatory behavior based on sensitive characteristics such as race and gender. One reason could be due to historical bias in the data." The thing about this discussion is that it aims to balance social welfare goals and profit maximizing goals but without any criteria for attaining basic fairness on a general level. To wit, suppose a machine learning algorithm is looking at twenty or fifty piece of data about a given individual, all of which are entirely irrelevant to the individual's chance of repaying a loan. But by random chance, one of those pieces of data, say handedness, happens to be correlated with a group's repayment history. So, some individuals with a good handedness are given loans more frequently and some individuals, with a bad handedness, are given loans less frequently. This situation doesn't matter to the company since the data is irrelevant and they only give out so many loans anyway and shitting on, say, left-handed people, gives them no grief. Moreover, if this trend is noted by all the companies soon it will be made "true". Which to say, companies making life-defining decisions like mortgages or parole-granting, just should be prohibited from the using a lasagna of random data to make their decisions and instead should be required to use specific rules with specific reasons behind them. And cry me a river about missed chances for optimization. This about the structure of society and optimization doesn't benefit society here imo.
- lambdaphagy 9y agoThis isn't about overfitting. Race is almost exactly the opposite problem: the signal in the data is so strong that you can't help but pick it up, even after you censor everything you can think of. If you try to predict any kind of consumer behavior from a racially heterogeneous dataset you will end up finding something that correlates with race, because practically everything correlates with race.
- mistermann 9y agoIs this actually true, that everything correlates with race? I know many things do, but does it even exceed 50%?
- deleted 9y ago
- ralusek 9y agoIf you use data such as repayment history, income, assets, and other metrics highly related to whether or not somebody is going to pay back a loan, and the resulting model does not output equal representation among arbitrary population groups, I fail to see a problem. If the data being used to train the model has absolutely no knowledge of arbitrary group identifiers, such as "gender" or "race," the resulting output quite simply not biased. However, if something like "race" or "gender" is actually being used as a feature input, then the output from most ML strategies is highly likely to pick up on correlations between racial and gender groups and certain outputs. That is undoubtedly going to lead to negative discriminatory outputs. So while I see absolutely no problem with the first scenario I mentioned, and I clearly see a problem with the second scenario...I'm inclined to believe that the politics of many people would lead them to have a problem with both.
- skybrian 9y agoThe thing that makes this trickier than it looks is that these features aren't necessarily encoded in a straightforward way. For example, race could be inferred with some loss of accuracy from zip code. Add more variables and it might be inferred with more accuracy. Machine learning doesn't necessarily do what you want it to do; it probably will "cheat" and use an easier proxy variable, unless you watch out for it. There is the urban legend about the machine learning program that learned to "recognize" tanks based on time of day in the training set. It probably never happened [1] but it gets the point across. [1] https://www.gwern.net/Tanks https://www.gwern.net/Tanks
- ralusek 9y agoYa, I agree with you that it's definitely tricky business. Including features like zipcode in something like a loan payment predictor is a good example of a sloppy proxy for the underlying and relevant features. Especially when certain data becomes hard to resist using, when something like zipcode of residence might have a 90% overlap with income range and be extremely trivial to acquire, whereas getting detailed and accurate income information from an applicant might be much harder. Tricky stuff.
- 9y ago