<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div></div><div><br></div><div>On Apr 7, 2016, at 1:01 PM, John Clark <<a href="mailto:johnkclark@gmail.com">johnkclark@gmail.com</a>> wrote:<br><div class="gmail_default" style="font-family: arial, helvetica, sans-serif; display: inline;"></div><br><div class="gmail_default" style="font-family: arial, helvetica, sans-serif; display: inline;"><font size="4">people would love to program a computer so that it has the best chance of solving the traveling salesman problem but nobody knows how to do it. </font></div> <br></div><div>I took a massive online course on a related topic where, in an early class, we learned Bayesian methods as applied to the traveling salesman problem. It was used as an example of code that could applied to machine learning. So I'm pretty sure this has been done with Bayesian and probably other methods. </div><div>-Henry</div></body></html>