Also: this supermodell example is utter bullshit. Enough men just like an ordinary woman and enough women just want an ordinary man.
The problem is that photos, generally, look less attractive than people do in real life, in part because the photo becomes such a focal point in contrast to real life where there is a whole person.
i wonder if the algorithm is smart enough to offset male attractiveness based on bank account balance
The paper's approach is based not on AI computer vision of attractiveness, nor even "hot or not" scoring by some audience, but rather upon who receives "initial contacts" and from whom. These initial contacts comprise the edges of a bipartite graph.
So if a man's profile photo includes his Porsche or Rolex, or he mentions in his text copy how much "he loves living in [swanky neighborhood]", then, yes, this approach would reflect attraction towards wealth.
How do the LEOs know what someone's intention is?
They only need probable cause, a threshold which can be less than 50-50 according to a 1983 Supreme Court decision.
From deep within the PDF (second link):
The two primary causes of increased fault rates at higher altitude are reduced cooling due to lower air pressure and increased cosmic ray-induced neutron strikes.
(Living in Colorado, I thought perhaps chips suffered from the same spurting newly opened toothpaste tube problem when manufactured at low altitude and installed into operation at high altitude, but it turned out the hypothesis was different, and, of course, left out of the Slashdot summary.)
That is why the concept of health insurance as business is inherently wrong. It is about threatening people with events out of their control into taking a bet which is unfavourable to them.
The insurance companies are not threatening people with illness; nature is.
Cutting out the extra tiers will cut costs immensely and save money by pure volume.
The waste of governmental corruption exceeds the savings from economies of scale.