Fractal Dimension Generalization Measure

TL;DR

This paper introduces a novel generalization measure for machine learning models based on fractal dimension analysis of decision boundaries, aiming to improve understanding of model performance.

Contribution

It proposes a new complexity measure using fractal dimension to predict generalization, advancing beyond existing boundary-focused methods.

Findings

Fractal dimension correlates with model generalization performance.

The measure was tested in the 'Predicting Generalization in Deep Learning' competition.

Results show improved prediction accuracy over traditional metrics.

Abstract

Developing a robust generalization measure for the performance of machine learning models is an important and challenging task. A lot of recent research in the area focuses on the model decision boundary when predicting generalization. In this paper, as part of the "Predicting Generalization in Deep Learning" competition, we analyse the complexity of decision boundaries using the concept of fractal dimension and develop a generalization measure based on that technique.