This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
Why would academia cancel the researchers? Is believing in X-risk from AI cancellable?
The academics doing AI ethics tend to be doing simpler things, e.g. trying to figure out what a "fair" lending algorithm is.
The technical challenge is finding an algo which spots hidden patterns that predict loan repayment except for the biggest pattern that predicts repayment (namely that blacks are much less likely to repay them, holding all else equal). But stating it in such explicit terms is a cancellable offense.
AND doing that without including the race variable during training.
The effect size is very strong, so it's pretty easy to find features correlated with race that capture it. One public graph I've seen is fig 7 in this paper which shows a 10-20% racial gap in non-delinquency (i.e. at a FICO score of 600, 40% of blacks and 20% of asians go delinquent for the particular loan product in that dataset).
More options
Context Copy link
If you train on all variables except race and black people are ceteris paribus less likely to repay, won't that just create a distinct cluster unexplained by any visible variables? Sounds simple enough to then take an average of all such clusters.
You pretty much need to include it as a variable and then 'correct' for it -- otherwise any half-decent AI will just route around its absence, as you suggest.
More options
Context Copy link
If you just leave race out of the input set, most likely the system will find some proxy for race which works, and your model will still show "bias". (A very strict reading of "ceteris paribus" would mean you couldn't find such a proxy, but that's not what is meant). If you leave race out of the input set, and train it with the goal of being "unbiased", you can get an "unbiased" predictor (that is inferior at prediction), but it's a little too obvious.
It kind of sounds like the whole discriminating against black people thing was a bright idea the AI hit upon when it was instructed not to discriminate against poor people.
More options
Context Copy link
More options
Context Copy link
Why would you do that if you want to make money on the loans you give?
Because the algorithms are not being written by greedy bankers.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link