Error Compensated Quantized SGD and its Applications to Large-scale Distributed Optimization
This addresses efficiency issues for practitioners in distributed machine learning, though it is incremental as it builds on existing quantization methods.
The paper tackles the communication bottleneck in large-scale distributed optimization by proposing an error compensated quantized SGD algorithm, which compresses gradients by up to 100 times without performance loss.
Large-scale distributed optimization is of great importance in various applications. For data-parallel based distributed learning, the inter-node gradient communication often becomes the performance bottleneck. In this paper, we propose the error compensated quantized stochastic gradient descent algorithm to improve the training efficiency. Local gradients are quantized to reduce the communication overhead, and accumulated quantization error is utilized to speed up the convergence. Furthermore, we present theoretical analysis on the convergence behaviour, and demonstrate its advantage over competitors. Extensive experiments indicate that our algorithm can compress gradients by a factor of up to two magnitudes without performance degradation.