Stochastic Gradient Descent with Hyperbolic-Tangent Decay on Classification
This work addresses the critical issue of learning rate scheduling for deep learning practitioners, offering an incremental improvement over existing schedulers.
The paper tackles the problem of learning rate scheduling in deep neural network training by proposing a new method called hyperbolic-tangent decay (HTD), which outperforms step decay and cosine schedulers on benchmarks like CIFAR-10, CIFAR-100, ImageNet, and Fashion-MNIST, requiring fewer hyperparameters and offering more flexibility.
Learning rate scheduler has been a critical issue in the deep neural network training. Several schedulers and methods have been proposed, including step decay scheduler, adaptive method, cosine scheduler and cyclical scheduler. This paper proposes a new scheduling method, named hyperbolic-tangent decay (HTD). We run experiments on several benchmarks such as: ResNet, Wide ResNet and DenseNet for CIFAR-10 and CIFAR-100 datasets, LSTM for PAMAP2 dataset, ResNet on ImageNet and Fashion-MNIST datasets. In our experiments, HTD outperforms step decay and cosine scheduler in nearly all cases, while requiring less hyperparameters than step decay, and more flexible than cosine scheduler. Code is available at https://github.com/BIGBALLON/HTD.