Recently, optimal time variable learning in deep neural networks (DNNs) was introduced in arXiv:2204.08528. 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.

翻译：近期，深度神经网络（DNNs）中的最优时间变量学习方法已在arXiv:2204.08528中提出。本文通过引入一个与离散动力系统时间范围直接相关的正则化项，对该概念进行了扩展。此外，我们提出了一种针对残差神经网络（ResNets）的自适应剪枝方法，该方法在降低网络复杂性的同时不牺牲表达能力，并同步缩短训练时间。通过将所提概念应用于MNIST和Fashion MNIST经典数据集上的分类任务，验证了其有效性。我们的PyTorch代码已开源在https://github.com/frederikkoehne/time_variable_learning。