QSGD: Communication-Efficient SGD via Gradient Quantization and Encoding
This addresses the communication bottleneck in large-scale parallel machine learning training, offering a provably convergent solution with practical speedups, though it is incremental as it builds on existing quantization heuristics.
The paper tackles the high communication cost in parallel stochastic gradient descent (SGD) by proposing QSGD, a family of gradient quantization and encoding schemes that provably converge, enabling trade-offs between compression and convergence time. It demonstrates significant reductions in communication cost and training time, such as training a ResNet-152 on ImageNet 1.8x faster on 16 GPUs while maintaining full accuracy.
Parallel implementations of stochastic gradient descent (SGD) have received significant research attention, thanks to excellent scalability properties of this algorithm, and to its efficiency in the context of training deep neural networks. A fundamental barrier for parallelizing large-scale SGD is the fact that the cost of communicating the gradient updates between nodes can be very large. Consequently, lossy compression heuristics have been proposed, by which nodes only communicate quantized gradients. Although effective in practice, these heuristics do not always provably converge, and it is not clear whether they are optimal. In this paper, we propose Quantized SGD (QSGD), a family of compression schemes which allow the compression of gradient updates at each node, while guaranteeing convergence under standard assumptions. QSGD allows the user to trade off compression and convergence time: it can communicate a sublinear number of bits per iteration in the model dimension, and can achieve asymptotically optimal communication cost. We complement our theoretical results with empirical data, showing that QSGD can significantly reduce communication cost, while being competitive with standard uncompressed techniques on a variety of real tasks. In particular, experiments show that gradient quantization applied to training of deep neural networks for image classification and automated speech recognition can lead to significant reductions in communication cost, and end-to-end training time. For instance, on 16 GPUs, we are able to train a ResNet-152 network on ImageNet 1.8x faster to full accuracy. Of note, we show that there exist generic parameter settings under which all known network architectures preserve or slightly improve their full accuracy when using quantization.