Highly Scalable Deep Learning Training System with Mixed-Precision: Training ImageNet in Four Minutes

Explore this paper's citation graph

Summary

This work builds a highly scalable deep learning training system for dense GPU clusters with three main contributions: a mixed-precision training method that significantly improves the training throughput of a single GPU without losing accuracy, an optimization approach for extremely large mini-batch size that can train CNN models on the ImageNet dataset without lost accuracy, and highly optimized all-reduce algorithms.

Type
preprint
Published
2018-07-30
Cited by
424
References
31
Access
Open access

Keywords

Training (meteorology), Scalability, Computer science, Artificial intelligence, Machine learning

References

Cited by

Related papers