Time Regularization in Optimal Time Variable Learning

Titelangaben

Herberg, Evelyn ; Herzog, Roland ; Köhne, Frederik:
Time Regularization in Optimal Time Variable Learning.
In: Proceedings in Applied Mathematics and Mechanics. Bd. 24 (2024) Heft 1 . - e202300299.
ISSN 1617-7061
DOI: https://doi.org/10.1002/pamm.202300299

Weitere URLs

Angaben zu Projekten

Abstract

Recently, optimal time variable learning in deep neural networks was introduced in Antil et al. In this manuscript we extend the concept by introducing a regularization term that directly relates to the time horizon in discrete dynamical systems. Furthermore, we propose an adaptive pruning approach for Residual Neural Networks (ResNets), which reduces network complexity without compromising expressiveness, while simultaneously decreasing training time. The results are illustrated by applying the proposed concepts to classification tasks on the well known MNIST and Fashion MNIST data sets. Our PyTorch code is available on https://github.com/frederikkoehne/time_variable_learning.