DeepReduce: A Sparse-tensor Communication Framework for Distributed Deep Learning

Abstract

Sparse tensors appear frequently in distributed deep learning, either as a direct artifact of the deep neural network's gradients, or as a result of an explicit sparsification process. Existing communication primitives are agnostic to the peculiarities of deep learning; consequently, they impose unnecessary communication overhead. This paper introduces DeepReduce, a versatile framework for the compressed communication of sparse tensors, tailored for distributed deep learning. DeepReduce decomposes sparse tensors in two sets, values and indices, and allows both independent and combined compression of these sets. We support a variety of common compressors, such as Deflate for values, or run-length encoding for indices. We also propose two novel compression schemes that achieve superior results: curve fitting-based for values and bloom filter-based for indices. DeepReduce is orthogonal to existing gradient sparsifiers and can be applied in conjunction with them, transparently to the end-user, to significantly lower the communication overhead. As proof of concept, we implement our approach on Tensorflow and PyTorch. Our experiments with large real models demonstrate that DeepReduce transmits fewer data and imposes lower computational overhead than existing methods, without affecting the training accuracy.

References in corpus (11)

• Language Models are Few-Shot Learners
• Exploring the Limits of Language Modeling
• Horovod: fast and easy distributed deep learning in TensorFlow
• Neural Collaborative Filtering
• Error Feedback Fixes SignSGD and other Gradient Compression Schemes
• Scaling Distributed Machine Learning with In-Network Aggregation
• Stochastic Distributed Learning with Gradient Quantization and Variance Reduction
• Double Quantization for Communication-Efficient Distributed Optimization
• Fast and Faster Convergence of SGD for Over-Parameterized Models and an Accelerated Perceptron
• rTop-k: A Statistical Estimation Approach to Distributed SGD
• Variance Reduction with Sparse Gradients