Automated Analysis of Experiments using Hierarchical Garrote

TL;DR

This paper introduces HiGarrote, an automated hierarchical variable selection method for experimental analysis that combines generalized ridge regression and nonnegative garrote, offering a fast, tuning-free approach that improves over existing methods.

Contribution

It presents HiGarrote, a novel hierarchical variable selection technique that integrates generalized ridge regression with nonnegative garrote, enhancing experimental analysis automation.

Findings

Demonstrates improved accuracy over existing methods

Shows faster analysis with no manual tuning required

Validates effectiveness on real experimental data

Abstract

In this work, we propose an automatic method for the analysis of experiments that incorporates hierarchical relationships between the experimental variables. We use a modified version of nonnegative garrote method for variable selection which can incorporate hierarchical relationships. The nonnegative garrote method requires a good initial estimate of the regression parameters for it to work well. To obtain the initial estimate, we use generalized ridge regression with the ridge parameters estimated from a Gaussian process prior placed on the underlying input-output relationship. The proposed method, called HiGarrote, is fast, easy to use, and requires no manual tuning. Analysis of several real experiments are presented to demonstrate its benefits over the existing methods.