Faster Neural Network Training with Approximate Tensor Operations
This addresses the computational bottleneck in training large neural networks, offering a practical speed-up for researchers and practitioners, though it is incremental as it builds on existing approximation methods.
The paper tackles the problem of slow deep neural network training by applying sample-based approximation to tensor operations, achieving up to 66% reduction in computations and 1.37x faster training time with negligible impact on test accuracy.
We propose a novel technique for faster deep neural network training which systematically applies sample-based approximation to the constituent tensor operations, i.e., matrix multiplications and convolutions. We introduce new sampling techniques, study their theoretical properties, and prove that they provide the same convergence guarantees when applied to SGD training. We apply approximate tensor operations to single and multi-node training of MLP and CNN networks on MNIST, CIFAR-10 and ImageNet datasets. We demonstrate up to 66% reduction in the amount of computations and communication, and up to 1.37x faster training time while maintaining negligible or no impact on the final test accuracy.