Happy new year Signalen devs!
My partner sent me a paper on making machine learning models transparent
and I expect those of you who tinker with training the ML models will also
find it good food for thought:
http://vanderschaar-lab.com/papers/NIPS2019_DBM.pdf
Jump ahead to section 5.2 "Predicting prognosis for breast cancer" to get
an idea about what they're doing.
This could be a very useful tool for combining machine learning with public
policy.
Pragmatically, it might allow Signalen to publish a transparent and
re-usable model without having to publish any sensitive data required for
generating a model.
At a glance, it seems there's more work on building models that model
models -- what a sentence! -- and "Explainable AI" seems to be a good
keyword to look up ... but I am very novice in this subject.
Cheers,
-Eric
--
Eric Herman, Lead codebase steward for quality
Foundation for Public Code |
https://publiccode.net
github.com/publiccodenet | github.com/ericherman
eric@publiccode.net | +31 620719662 | @Eric_Herman
------------------------------
*From:* Medlock, S.K.
s.k.medlock@amsterdamumc.nl
*Sent:* Wednesday, January 15, 2020 14:32
*Subject:* interesting work from Mihaela van der Schaar
Just heard a talk from one of the machine learning people in the UK. She
had a number of interesting topics, but this is the one that caught my
attention:
http://vanderschaar-lab.com/papers/NIPS2019_DBM.pdf
Figured it might interest you too.
Stephanie "Ace" Medlock
Assistant Professor | Amsterdam UMC - Location AMC
Department of Medical Informatics | Meibergdreef 15, 1105AZ Amsterdam
www.amsterdamumc.nl
______________________________________________________
VUmc disclaimer : www.vumc.nl/disclaimer
AMC disclaimer : www.amc.nl/disclaimer