Skip to main content

and
  1. No Access

    Chapter and Conference Paper

    Treating Artificial Neural Net Training as a Nonsmooth Global Optimization Problem

    We attack the classical neural network training problem by successive piecewise linearization, applying three different methods for the global optimization of the local piecewise linear models. The methods are...

    Andreas Griewank, Ángel Rojas in Machine Learning, Optimization, and Data Science (2019)