3 ms·

Maybe there is hope for 2020

2 points by jjgreen 6y ago

jjgreen 6y ago
Full title: Maybe there is hope for 2020: AI that 'predicts criminality' from faces with '80% accuracy, no bias' gets in the sea; Springer ditches paper from research tome after mass protest by boffins over junk science
- dredmorbius 6y ago
  2nd part of title would be preferable to the 1st which you've used here.
  - jjgreen 6y ago
    Just trying to follow the site guidelines, The Reg' makes that hard, hence the explanatory comment ...
    - dredmorbius 6y ago
      There's some interpretation allowed, even encouraged. The goal seems to be to stick with something from the source rather than creating an entirely novel title. If the title begins with a number or number + gratuitous adjective, we'd appreciate it if you'd crop it. E.g. translate "10 Ways To Do X" to "How To Do X," and "14 Amazing Ys" to "Ys." Exception: when the number is meaningful, e.g. "The 5 Platonic Solids." Otherwise please use the original title, unless it is misleading or linkbait; don't editorialize. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html General guidance, from dang, 2015: https://news.ycombinator.com/item?id=10597136 https://news.ycombinator.com/item?id=10597136 Reg example: https://news.ycombinator.com/item?id=21277129 https://news.ycombinator.com/item?id=21277129 Long title rewrite: https://news.ycombinator.com/item?id=16662989 https://news.ycombinator.com/item?id=16662989
      - jjgreen 6y ago
        Ah thanks!
jaekash 6y ago
Flagged to fix the title. This is the letter that supposedly called it junk science: https://medium.com/@CoalitionForCriticalTechnology/abolish-the-techtoprisonpipeline-9b5b14366b16 https://medium.com/@CoalitionForCriticalTechnology/abolish-t... The central claim that the title of the original paper would suggest is that criminality can be predicted by using image processing (original title: "A Deep Neural Network Model to Predict Criminality Using Image Processing"). This is the press release of the original paper: https://web.archive.org/web/20200506013352/https://harrisburgu.edu/hu-facial-recognition-software-identifies-potential-criminals/ https://web.archive.org/web/20200506013352/https://harrisbur... The response letter claims: > AI applications that claim to predict criminality based on physical characteristics are a part of a legacy of long-discredited pseudosciences such as physiognomy and phrenology, which were and are used by academics, law enforcement specialists, and politicians to advocate for oppressive policing and prosecutorial tactics in poor and racialized communities. The thing is though, either the claim of the paper is right or wrong, if they can predict criminality with 80% accuracy from faces based on image processing then it can do that. So either they can, or they cannot. If they cannot, then challenge them on that aspect of their paper, if they can, then no amount of "debunking" changes it. This seems a bit like some moral panic rather than a well founded scientific objection. If the claim is true, would it not be better to know the truth? Why should we hide claims about fact? If the claim about fact is wrong then it is wrong, no need to hide it, just say it is wrong. If it is right it is right, I really don't get the need for moral outrage here.

2 points by jjgreen 6y ago

Full title: Maybe there is hope for 2020: AI that 'predicts criminality' from faces with '80% accuracy, no bias' gets in the sea; Springer ditches paper from research tome after mass protest by boffins over junk science

dredmorbius 6y ago
2nd part of title would be preferable to the 1st which you've used here.
- jjgreen 6y ago
  Just trying to follow the site guidelines, The Reg' makes that hard, hence the explanatory comment ...
  - dredmorbius 6y ago
    There's some interpretation allowed, even encouraged. The goal seems to be to stick with something from the source rather than creating an entirely novel title. If the title begins with a number or number + gratuitous adjective, we'd appreciate it if you'd crop it. E.g. translate "10 Ways To Do X" to "How To Do X," and "14 Amazing Ys" to "Ys." Exception: when the number is meaningful, e.g. "The 5 Platonic Solids." Otherwise please use the original title, unless it is misleading or linkbait; don't editorialize. https://news.ycombinator.com/newsguidelines.html https://news.ycombinator.com/newsguidelines.html General guidance, from dang, 2015: https://news.ycombinator.com/item?id=10597136 https://news.ycombinator.com/item?id=10597136 Reg example: https://news.ycombinator.com/item?id=21277129 https://news.ycombinator.com/item?id=21277129 Long title rewrite: https://news.ycombinator.com/item?id=16662989 https://news.ycombinator.com/item?id=16662989
    - jjgreen 6y ago
      Ah thanks!

jaekash 6y ago

Flagged to fix the title. This is the letter that supposedly called it junk science: https://medium.com/@CoalitionForCriticalTechnology/abolish-the-techtoprisonpipeline-9b5b14366b16 https://medium.com/@CoalitionForCriticalTechnology/abolish-t... The central claim that the title of the original paper would suggest is that criminality can be predicted by using image processing (original title: "A Deep Neural Network Model to Predict Criminality Using Image Processing"). This is the press release of the original paper: https://web.archive.org/web/20200506013352/https://harrisburgu.edu/hu-facial-recognition-software-identifies-potential-criminals/ https://web.archive.org/web/20200506013352/https://harrisbur... The response letter claims: > AI applications that claim to predict criminality based on physical characteristics are a part of a legacy of long-discredited pseudosciences such as physiognomy and phrenology, which were and are used by academics, law enforcement specialists, and politicians to advocate for oppressive policing and prosecutorial tactics in poor and racialized communities. The thing is though, either the claim of the paper is right or wrong, if they can predict criminality with 80% accuracy from faces based on image processing then it can do that. So either they can, or they cannot. If they cannot, then challenge them on that aspect of their paper, if they can, then no amount of "debunking" changes it. This seems a bit like some moral panic rather than a well founded scientific objection. If the claim is true, would it not be better to know the truth? Why should we hide claims about fact? If the claim about fact is wrong then it is wrong, no need to hide it, just say it is wrong. If it is right it is right, I really don't get the need for moral outrage here.