2020/08/26 by Hanlin Tang, Shaoduo Gan, Tang, Hanlin +11 · 1 citation
Computer Science · Engineering · #Algorithms and Data Compression #Distributed #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Parallel #Robotic Path Planning Algorithms #Vehicle License Plate Recognition #and Cluster Computing (cs.DC)
paper · pdf · doi:10.48550/arxiv.2008.11343
openalex publication_date 2020/08/26 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Adam is the important optimization algorithm to guarantee efficiency and accuracy for training many important tasks such as BERT and ImageNet. However, Adam is generally not compatible with information (gradient) compression technology. Therefore, the communication usually becomes the bottleneck for parallelizing Adam. In this paper, we propose a communication efficient \bf ADAM \bf preconditioned \bf Momentum SGD algorithm-- named APMSqueeze-- through an error compensated method compressing gradients. The proposed algorithm achieves a similar convergence efficiency to Adam in term of epochs, but significantly reduces the running time per epoch. In terms of end-to-end performance (including the full-precision pre-condition step), APMSqueeze is able to provide sometimes by up to 2-10× speed-up depending on network bandwidth. We also conduct theoretical analysis on the convergence and efficiency.