Accelerating DNN Training with Structured Data Gradient Pruning
This addresses the need for faster DNN training for practitioners, but it is incremental as it builds on existing pruning techniques with hardware-specific optimizations.
The paper tackles the problem that weight pruning typically does not speed up DNN training and can slow convergence, proposing Structured Data Gradient Pruning (SDGP) to achieve a 15-25% reduction in total training time without significant performance impact.
Weight pruning is a technique to make Deep Neural Network (DNN) inference more computationally efficient by reducing the number of model parameters over the course of training. However, most weight pruning techniques generally does not speed up DNN training and can even require more iterations to reach model convergence. In this work, we propose a novel Structured Data Gradient Pruning (SDGP) method that can speed up training without impacting model convergence. This approach enforces a specific sparsity structure, where only N out of every M elements in a matrix can be nonzero, making it amenable to hardware acceleration. Modern accelerators such as the Nvidia A100 GPU support this type of structured sparsity for 2 nonzeros per 4 elements in a reduction. Assuming hardware support for 2:4 sparsity, our approach can achieve a 15-25\% reduction in total training time without significant impact to performance. Source code and pre-trained models are available at \url{https://github.com/BradMcDanel/sdgp}.