Slashdot Mirror
Google Research Promotes Equality In Machine Learning, Doesn't Mention Age
An anonymous reader writes: New research from Google Brain examines the problem of 'prejudice by inference' in supervised learning -- the syndrome by which 'fairness through unawareness' can fail; for example, when the information that a loan applicant is female is not included in the data set, but gender can be inferred from other data factors which are included, such as whether the applicant is a single parent. Since 82% of single parents are female, there is a high probability that the applicant is female. The proposed framework shifts the cost of poor predictions to the decision-maker, who is responsible for investing in the accuracy of their prediction systems. Though Google Brain's proposals aim to reduce or eliminate inadvertent prejudice on the basis of race, religion or gender, it is interesting to note that it makes no mention of age prejudice -- currently a subject of some interest to Google.
I think that AIs, by definition, cannot have bias. Or rather, they cannot be bigoted or prejudiced. They have entirely fair and generally accurate views of groups. It's like charging 16 year olds more for auto insurance is justified, just on a larger scale.
Please don't build affirmative action into machine learning. If someone is higher risk for something and we want to be unfair and charge them the same rate as a person with low risk - we can sort that out at a higher layer. Don't mess with the raw data.
"At the heart of our approach is the idea that individuals who qualify for a desirable outcome should have an equal chance of being correctly classified for this outcome."
Sounds like they need a random number generator, not a machine learning algorithm.
"Promotes equality" being a euphemism for "deliberately stupefy". Another example of Tay's Law.
If you read the paper, by forcing the algorithm to be "fair" their accuracy and hypothetical profit goes down.
Why isn't Hillary in jail?
to make decisions, the system will always be sexist and racist. That data will indicate they're a riskier loan, so even if it isn't intended to destroy the lives of women and minorities, it will destroy their lives.
The fact that no matter how good it is, google 'brain' can't strictly speaking, 'infer' anything. Good grief, they are stupid.
Can somebody explain to me how this is a good idea? How is less information for decisions a good thing?
Don't know about everybody else, but when I am making decisions, I like to have the most information possible.
So if women are 2x as likely to default as men on a loan (MADE UP NUMBER, NOT BASED ON FACTS), damn right that is important to know when considering to give the loan and at what interest rate. This would not be sexism or bigotry or wtv else regressive fascist femenazis and SJW would have you believe. It would be an important variable when measuring risk.
When a loan application cannot collect that information, be it direct or inferred, well quite simply EVERYBODY will pay more in order to guarantee the lender does not get screwed in the end.
If the default rates of men and women are the same, then that variable is of no consequence and nobody would bother asking or wasting time to infer it. Unless of course you will try to make the argument that there are loan managers out there willing to lose their job/raise/bonus/promotion in order to deny women loans and not meet his targets.
WTF guys? Why isn't anyone having the right conversations?
Hard AI is still in its infancy (and in my opinion very well may never happen with silicon computers as we know it).
There are much simpler problems to solve than trying to have an AI lie to itself (and others) that the difference it sees is not actually a difference. How do you even begin to explain political correctness to an AI?
Hey AI, instead of selection the best option(s) under the criteria you were given, you need to make an inferior choice, because... well, reasons...
Hey AI, you need to see the legally protected differences, and then later deny seeing them with some sort of parallel construction argument.
What utter nonsense.
If even machines come up with measurable differences between work performance of males and females, then I think giving them in average the same amount of money or the same promotions is discrimination. I'm all for giving a woman who performs just as well as a man the same money, but if there are additional risk factors like a pregnancy or when the parent has to raise child, the person usually prioritizes these things over work, so why should work not be allowed to prioritize that person over others who do not raise children or do not drop out for weeks and months out of some work-external reason.
Exactly. Single women defaulted at four times the rate IIRC from when I worked as a developer for a regional bank. Any system that reduces risk is inherently biased.
We can use the extra money to subsidise men's insurance premiums. Clearly, "prejudice by inference" is causing men to be charged too much.
I'm sure supporters of gender equality will agree with me.
For the example used, e.g. regarding getting a loan, how is 'single parent' actually relevant? Seriously. While there may be statistics that somehow demonstrate some relationship between 'single parent' and 'loan worthiness'. Is it not really more relevant on a specific individuals basis to know what their income is, how they get it and what their costs are? To me the assumption appears to be that a '2 parent home' is somehow more legally stable in terms of allocating income & costs, but given the nature of society, high rates of divorce etc, the idea that a '2 parent home' will remain so or that any individual in that '2 parent home' is 'carrying their weight' (allocating their income as might be 'expected') just doesn't seem valid.
The point I'm trying to make is that given the legal framework & lack of 'enforcement ability' to guarantee costs of child care are enforced, it wouldn't seem to matter to whether or not a given individual is capable of paying off a loan. It would seem more valid to know the specific legal details of where income is coming from, the ability guarantee that income & similarly for a persons costs.
Using statistics for predictive value in this case seems like a 'lazy way out' that simply continues to perpetuate a society that doesn't enforce proper cost allocation of child care (e.g. if there was a contract for child support that the bank relied on in regards to treating it as income & one of the people in that contract failed to meet the contractual requirements it seems the bank would have some stake in enforcing those provisions).
Having your property searched (trespassed on by police) is different than not getting a loan. You own your house. You don't own the bank's money.
...) try to evaluate risks as best they can. If you make them blind to a signal, but they are unwilling to increase their risk tolerance, they will behave more conservatively, not less. They will decrease their service and use even cruder methods to control their risk.
If police were not a privileged monopoly, they would owe restitution for bad searches, just like a trespasser does. But given that it is a monopoly, we try to rein its power in with rules.
The idea that the world is better or more rational by ignoring rational inferences is mistaken. Take for example the effort to "ban the box" (which means employers don't get to ask if you're a felon). Although such legislation are intended to help black people, but the the results appear to have been opposite [1].
People (including employers, creditors, insurers, retailers,
[1] http://phys.org/news/2016-06-e...
These comments are mine; I do not speak for my employer.
I love it. They are already fucking up their "artificial intelligence" (neural network) to be a liberal SJW.
The only "solution" will be if every living thing has the same result, so just ignore all values and hardcode the one output.
Maybe we can google bomb the fact that Google treats people over 30 with less dignity than Logan's Run.
So g00gle found out that different groups really are different in a number of relevant factors and their conclusion was that evil cisgendered bigots when seeing inferior relevant attributes are going to automatically figure out an applicant is a protected minority and in their mind are somehow going to skip over the relevant reasons to discriminate and solely discriminate against based on them being a minority and the effect will somehow be distinguishable and worse than if they had just stuck with discriminating with the relevant reasons they already have.
People are not equal. Get over it.
Life is not fair. Get over it.
I realize that you effeminate silicon valley one-worlders want so badly to deny the reality that is right in front of you but you can't. Get over it.
the machine learning algorithm infers a difference which is real, but uncomfortable for us socially.
Let's assume that we can prove that the detected difference was in the case NOT introduced by human-created input-data bias.
I'll give an example: I'm left handed so I think I'm allowed to talk about this.
What if the system learns that left handed people in North America die a little earlier than right handed people.
And specifically that they die with higher frequency in car accidents.
(I'm pretty sure both statements are true above. Reasons for it are not definite, but for the first one, can include that many tools and affordances in society are designed to be easy for right-handers, so left-handers may interact poorly with them sometimes and sometimes that bites, For the second, it may be because a left handed driver who dozes off or becomes distracted tends to pull the steering wheel a little to the left, into oncoming traffic, Right-handers tend to pull to the right, onto the on-balance safer shoulder of the road.)
So does that mean its ok to increase life insurance premiums and automobile insurance premiums for left-handed people?
What kind of statistically valid discrimination IS ok? Any?
Then what do we do, in this day and age?
Where are we going and why are we in a handbasket?
That's what I have learned about bigotry recently. I grew up thinking that bigotry was applying a conclusion to someone's behavior or outcome, which would only be true, if self reinforcing. But now being a bigot counts when applying a bias against a protected group, even if backed up research and data.
infer: "deduce or conclude (information) from evidence and reasoning rather than from explicit statements"
Can I infer that you haven't read much of the last 50 years' research literature in AI, formal logic, Bayesian inference, and machine learning?
Where are we going and why are we in a handbasket?
The government explicitly discriminates and even mandates the discrimination in law in numerous cases. The government utilizes theft and violence against people who don't comply too.
Examples:
Prohibits employers from hiring minors
Prohibits people under a certain age from driving and mandates permissions to do so (drivers licenses)
Requires persons under a certain age to attend school
etc.
I don't believe in government mandated age discrimination and think people who are against the use of violence to achieve political goals should join me in New Hampshire. The Free State Project is all about increasing freedom and liberty in our life time by joining like-minded people to one place. That place is New Hampshire. There is no place in NH today that you can move and not find other activists taking aim at the state. Check out: http://www.freestateproject.org/ http://www.freekeene.com/ http://www.shiresociety.com/
The problem Google is describing isn't limited to a subset of arbitrary tribal factors society deems to be off limits.
Entire reason for existence of these systems is making prejudiced decisions about individuals based on statistical evidence.
You can spend all day filtering out things that will get you sued or attract bad press but this doesn't address core fact these systems are intended to make prejudiced judgments about individuals based on statistical experience and evidence.
Being prejudiced can be practically helpful in some contexts but don't pretend that isn't what your doing, don't confuse it for fairness and don't bother making up a bunch of mystical bullshit about how your dataset or programmers are biased. Prejudice is the raison d'etre of these systems. It is what they are designed to do.
How can there be "prejudice" if the system _does not have cognition_? It just approximates a function. If a woman is less (or more likely) to default on a loan, it'll just say so, SJWs be damned. That's why women see ads for shoes even if they never disclosed that they are women to Google. That's also why they see fewer ads for engineering positions (women are statistically much less likely to be interested in engineering fields).
It's a function approximation problem, and this happens to be the function that the real world data seems to support. Now you want to wreck it for some kind of affirmative action, thus decreasing its accuracy and driving an agenda of what you think the world should look like, rather than what it actually is.
It can be, but the concept of "gender" or "race" is meaningless to a machine learning system for loan evaluations, and it has no biases or prejudices. If a properly trained machine learning system disproportionately rejects applications of some gender or race, then that reflects an actual statistical regularity in the world, not the result of discrimination or bias. Furthermore, if you force that system to make decisions that are representative of national demographics, it will make suboptimal decisions. The Google paper actually points this out. What they do is provide a method that allows for some degree of discrimination, but even their system is still suboptimal.
Yes, there are big statistical differences between different genders and racial groups in their propensity to commit violence, commit crimes, and repay loans. And these differences are increasing rather than decreasing because politics currently encourages a "multicultural society" and cultures differ enormously in a lot of areas.
There are some systems that are so complex (people going about their lives and having a chance of dying, for example) that you will never be able to predict the particular outcome for a particular individual, no matter if your computer brain is the size of a planet.
The best info we can ever get in advance about these complex systems is statistics about populations of the with similar characteristics in similar environments.
Where are we going and why are we in a handbasket?
was brought to you by
the association of resource-extraction-company security goons and the national henchmen's association.
Where are we going and why are we in a handbasket?
Gonna protect the cave.
Where are we going and why are we in a handbasket?
Let's say we put all available data in, sort out the crap data so the input is neutral.
Then we get exactly the prejudices out. This confirms them. Period.
This does not imply, that we should support them. This only implies, that they are there. People often jump to conclusions, that this implies causation, while it implies correlation. If some places have higher crime rate and some places have more black people there (another case of ML prejudices) and the data is correct, it's the correct decision for an insurance to raise the rates at these places. Because they can expect more cases.
This is the point, where exactly the people who are upset by the result from the data need to act. And change the circumstances.
For example maybe the blacks move away and the crime rate stays the same, but the black people who were associated (by the upset people misinterpreting the statistics) with the crime now live in a peaceful place with cheap insurance rates.
So the only thing it says is: You need to interpret statistics. Data doesn't lie, but your fast conclusions do.
Not really. Bigotry is misused instead of racism. Bigotry is not changing your views when facts tell you to. The majority of people seeking social justice are bigots while calling others the same because the data says something different from their message.
prejudice: preconceived opinion that is not based on reason or actual experience.
Of course, I used google to find that definition, so
BURN THE WITCH!
Bigotry in general is more about the systems that society has in place that combine to make it so that people with certain backgrounds are disadvantaged with respect to others. These systems are extremely varied and reinforced by a variety of societal traditions, personal prejudices, business practices, government practices, and more.
At an individual level, bigotry involves supporting and continuing those systems of oppression, whether consciously or unconsciously.
How can you have so many IT savy people in one place, and somehow everyone has drank the marking 'AI' koolaid?
Really?
Over and over, and over, thousands of posts talking about computer's sorting data, or "machine learning", as 'if it was "intelligent"'.
It is not frigen magic, in spite of your AI pornographic fantasies, when a computer sorts data. It is no more magic than a computer doing basic math faster than a human.
Yet, day after day, Slashdot is filled with morons sucking up the AI fantasy spin being sold them.
Bigotry in general is more about the systems that society has in place that combine to make it so that people with certain backgrounds are disadvantaged with respect to others. These systems are extremely varied and reinforced by a variety of societal traditions, personal prejudices, business practices, government practices, and more.
At an individual level, bigotry involves supporting and continuing those systems of oppression, whether consciously or unconsciously.
I will agree with that. But sometimes it feels like in the effort to remove bigotry (which I'm all for), some legitimate differences between groups of people (which aren't in place due to society) are getting covered over, even to our detriment.
hahahaha. Oh, wait. You're serious. Let me laugh even harder. HAHAHAHAHAHA.
What they are really saying is that learning machines are confirming politically incorrect beliefs. A lot of stereotypes are based on a kernel of truth, and given enough processing power and data that truth is coming to the forefront. When people were crunching the numbers is was easy to blame prejudice or some kind of *ism. But learning algorithms don't have that, they just learn patterns. What there researchers are doing has nothing to do with fostering equality, it's about avoiding embarrassing truths.
It reminds me of when polar explorers were shocked at the "sexual depravity "of penguins so they wrote their reports in Greek and kept the truth hidden. Sometimes society just isn't ready to handle the truth.
http://www.philica.com/display_observation.php?observation_id=135
Ukkonen, T. (2016). How to prevent discrimination of machine learning models against variables such as age or gender etc. ?. PHILICA.COM Observation number 135.
There is generally far more variation within groups of people than between them, though. For the most part, measured differences between different groups have proven to be due to research that didn't fully account for researchers' and society's biases.
Simple example: there's a stereotype that girls are bad at math. It's been demonstrated that merely reminding girls of the existence of that stereotype causes them to do worse on math tests. This is an example of stereotype threat, where the existence of the stereotype itself causes a cognitive burden: even knowing that the stereotype is bullshit doesn't prevent it from causing harm. You can bring girls' math scores back up by creating an environment where the stereotype is minimized. And, of course, if that stereotype is enforced during school for a few years, those girls will end up definitely worse at math than their male peers just because later math builds on earlier math.
So in essence, you can't be sure that most any measured difference between two groups of people is a real difference, rather than just a difference imposed by society.