Page 1 of 1

Evaluation doubt

Posted: Sat Oct 29, 2016 5:04 pm
by Fabio Gobbato
In the evaluation function is it better to have a few weights but well tuned or more weights not well tuned?
What does your experience say?

Re: Evaluation doubt

Posted: Sat Oct 29, 2016 5:30 pm
by ZirconiumX
Fewer, but better. Fewer weights helps with orthogonality, and the tuning may well make up for the missing terms. Plus, the terms will harmonise themselves with your search, increasing search speed.

Re: Evaluation doubt

Posted: Sun Oct 30, 2016 6:24 am
by BeyondCritics
Read chapter 5 of this book
http://www.deeplearningbook.org/
anf you can dicuss the question yourself. Or read any introduction into "Machine Learning".

Re: Evaluation doubt

Posted: Mon Oct 31, 2016 12:52 pm
by Ferdy
BeyondCritics wrote:Read chapter 5 of this book
http://www.deeplearningbook.org/
anf you can dicuss the question yourself. Or read any introduction into "Machine Learning".
Good info, thanks.

Re: Evaluation doubt

Posted: Mon Oct 31, 2016 7:20 pm
by Fabio Gobbato
Of course less weights are easier to tune, but even if they are well tuned they can't express all the knowledge of a more complex evaluation.
So the answer is not so easy.