Globally Optimal MINLP Formulation for Symbolic Regression

We describe a mathematical program which finds an expression tree of least description complexity that fits a set of observations as well as required.

By: Lior Horesh, Leo Liberti, Haim Avron

Published in: RC25620 in 2016

rc25620.pdf

Questions about this service can be mailed to reports@us.ibm.com .