Equations are not being displayed properly on some articles. We hope to have this fixed soon. Our apologies.

Ukkonen, T. (2016). How to prevent discrimination of machine learning models against variables such as age or gender etc. ?. PHILICA.COM Observation number 135.

ISSN 1751-3030  
Log in  
Register  
  1187 Articles and Observations available | Content last updated 23 September, 10:52  
Philica entries accessed 3 311 929 times  


NEWS: The SOAP Project, in collaboration with CERN, are conducting a survey on open-access publishing. Please take a moment to give them your views

Submit an Article or Observation

We aim to suit all browsers, but recommend Firefox particularly:

How to prevent discrimination of machine learning models against variables such as age or gender etc. ?

Tomas Ukkonenunconfirmed user (Helsinki University of Technology)

Published in matho.philica.com

Observation
It is possible to integrate over nuisance variables n:

Integral p(loan, n | input) dn =
Integral p(loan | n, input) p(n | input) dn

And it is possible to model machine learning by modelling them as probability distributions. For example, in case of gender one can create specific models for each gender and calculate:

p(loan | input) =

p(loan | M, input) p(M | input) + p(loan | F, input) p(F | input)

Information about this Observation
This Observation has not yet been peer-reviewed
This Observation was published on 12th October, 2016 at 17:58:14 and has been viewed 1149 times.

Creative Commons License
This work is licensed under a Creative Commons Attribution 2.5 License.
The full citation for this Observation is:
Ukkonen, T. (2016). How to prevent discrimination of machine learning models against variables such as age or gender etc. ?. PHILICA.COM Observation number 135.


<< Go back Review this ObservationPrinter-friendlyReport this Observation



Website copyright © 2006-07 Philica; authors retain the rights to their work under this Creative Commons License and reviews are copyleft under the GNU free documentation license.
Using this site indicates acceptance of our Terms and Conditions.

This page was generated in 0.4310 seconds.