## Title data

Herberg, Evelyn ; Herzog, Roland ; Köhne, Frederik:

**Time Regularization in Optimal Time Variable Learning.**

Heidelberg
,
2023
. - 9 p.

DOI: https://doi.org/10.48550/arXiv.2306.16111

## Project information

Project title: |
Project's official title Project's id Multilevel Architectures and Algorithms in Deep Learning 464103607 |
---|---|

Project financing: |
Deutsche Forschungsgemeinschaft |

## Abstract in another language

Recently, optimal time variable learning in deep neural networks (DNNs) was introduced in Antil, Díaz, Herberg, 2022. 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, Köhne, 2023.

## Further data

Item Type: | Preprint, postprint |
---|---|

Refereed: | Yes |

Keywords: | deep learning; deep neural networks; network architecture; PyTorch |

Institutions of the University: | Faculties Faculties > Faculty of Mathematics, Physics und Computer Science Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics > Chair Applied Mathematics Faculties > Faculty of Mathematics, Physics und Computer Science > Department of Mathematics > Chair Applied Mathematics > Chair Applied Mathematics - Univ.-Prof. Dr. Anton Schiela |

Result of work at the UBT: | Yes |

DDC Subjects: | 500 Science > 510 Mathematics |

Date Deposited: | 03 Jul 2023 09:18 |

Last Modified: | 03 Jul 2023 10:11 |

URI: | https://eref.uni-bayreuth.de/id/eprint/85900 |