Saved in:
Bibliographic Details
Main Authors: Datres, Massimiliano, Leonardi, Gian Paolo, Figalli, Alessio, Sutter, David
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2401.09184
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909670681083904
author Datres, Massimiliano
Leonardi, Gian Paolo
Figalli, Alessio
Sutter, David
author_facet Datres, Massimiliano
Leonardi, Gian Paolo
Figalli, Alessio
Sutter, David
contents We introduce a novel capacity measure 2sED for statistical models based on the effective dimension. The new quantity provably bounds the generalization error under mild assumptions on the model. Furthermore, simulations on standard data sets and popular model architectures show that 2sED correlates well with the training error. For Markovian models, we show how to efficiently approximate 2sED from below through a layerwise iterative approach, which allows us to tackle deep learning models with a large number of parameters. Simulation results suggest that the approximation is good for different prominent models and data sets.
format Preprint
id arxiv_https___arxiv_org_abs_2401_09184
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Two-Scale Complexity Measure for Deep Learning Models
Datres, Massimiliano
Leonardi, Gian Paolo
Figalli, Alessio
Sutter, David
Machine Learning
We introduce a novel capacity measure 2sED for statistical models based on the effective dimension. The new quantity provably bounds the generalization error under mild assumptions on the model. Furthermore, simulations on standard data sets and popular model architectures show that 2sED correlates well with the training error. For Markovian models, we show how to efficiently approximate 2sED from below through a layerwise iterative approach, which allows us to tackle deep learning models with a large number of parameters. Simulation results suggest that the approximation is good for different prominent models and data sets.
title A Two-Scale Complexity Measure for Deep Learning Models
topic Machine Learning
url https://arxiv.org/abs/2401.09184