SERVING THE QUANTITATIVE FINANCE COMMUNITY

 
User avatar
katastrofa
Posts: 8036
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: Hacking ML for fun & profit

March 17th, 2019, 11:56 am

That was my point. Equating gradient methods to simulated annealing (artificial time vs artificial temperature) is not a far jump, but in the wrong dimension. Physicists developed those methods and name things like that for a reason ;-)
 
User avatar
Cuchulainn
Posts: 59665
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: Hacking ML for fun & profit

March 17th, 2019, 12:41 pm

That was my point. Equating gradient methods to simulated annealing (artificial time vs artificial temperature) is not a far jump, but in the wrong dimension. Physicists developed those methods and name things like that for a reason ;-)
Maybe the variable [$]t[$] causes confusion. Better to say [$]T = T(t)[$] and then you get a double whammy. A kind of 1:1 correspondence.
 
User avatar
katastrofa
Posts: 8036
Joined: August 16th, 2007, 5:36 am
Location: Alpha Centauri

Re: Hacking ML for fun & profit

March 17th, 2019, 1:54 pm

  *groans*
Image
 
User avatar
Cuchulainn
Posts: 59665
Joined: July 16th, 2004, 7:38 am
Location: Amsterdam
Contact:

Re: Hacking ML for fun & profit

March 17th, 2019, 3:56 pm

That's not a groan, it's a snarl :D
ABOUT WILMOTT

PW by JB

Wilmott.com has been "Serving the Quantitative Finance Community" since 2001. Continued...


Twitter LinkedIn Instagram

JOBS BOARD

JOBS BOARD

Looking for a quant job, risk, algo trading,...? Browse jobs here...


GZIP: On