Large Batch Training of Convolutional Networks
This addresses optimization difficulties in distributed deep learning for researchers and practitioners, offering a more general solution than existing methods.
The paper tackled the problem of training large convolutional networks with large batch sizes, which often leads to lower model accuracy, by proposing a new training algorithm based on Layer-wise Adaptive Rate Scaling (LARS), enabling scaling AlexNet to a batch size of 8K and ResNet-50 to 32K without accuracy loss.
A common way to speed up training of large convolutional networks is to add computational units. Training is then performed using data-parallel synchronous Stochastic Gradient Descent (SGD) with mini-batch divided between computational units. With an increase in the number of nodes, the batch size grows. But training with large batch size often results in the lower model accuracy. We argue that the current recipe for large batch training (linear learning rate scaling with warm-up) is not general enough and training may diverge. To overcome this optimization difficulties we propose a new training algorithm based on Layer-wise Adaptive Rate Scaling (LARS). Using LARS, we scaled Alexnet up to a batch size of 8K, and Resnet-50 to a batch size of 32K without loss in accuracy.