ML and CH

[JOSEPH SHIPMAN]

This paper shows that a combinatorial problem with an interpretation in the field of machine learning is equivalent to the continuum hypothesis.

https://arxiv.org/pdf/1711.05195.pdf

I expect there is a reformulation of the machine learning situation that is empirically indistinguishable from it but which corresponds to combinatorics decidable within ZF; but I’d like to hear what computer scientists have to say about this.

— JS

Sent from my iPhone
-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/fom/attachments/20210224/ead9aa27/attachment-0001.html>