Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
Hyperparameter optimization
open-in-new
<p>In <a href="/facts/Machine_learning/e0w0XJTu">machine learning</a>, hyperparameter optimization or tuning is the problem of choosing a set of optimal <a href="/facts/Hyperparameter_(machine_learning)/jCxm3fTA">hyperparameters</a> for a learning algorithm. A hyperparameter is a <a href="/facts/Parameter/zFvVFMv9">parameter</a> whose value is used to control the learning process, which must be configured before the process starts. </p><p>Hyperparameter optimization determines the set of hyperparameters that yields an optimal model which minimizes a predefined <a href="/facts/Loss_function/xv5ozuhl">loss function</a> on a given <a href="/facts/Data_set/eDvegUTK">data set</a>. The objective function takes a set of hyperparameters and returns the associated loss. <a href="/facts/Cross-validation_(statistics)/pDWg8s0e">Cross-validation</a> is often used to estimate this generalization performance, and therefore choose the set of values for hyperparameters that maximize it. </p>