arXiv (Cornell University) · 2018 · 78 citations · 36 references
ComputingComputer EngineeringComputer ScienceMachine LearningParallel AlgorithmsParallel ComputingData ParallelismDeep LearningGraphics Processing Unit ArchitectureGraphics Processing Unit ComputingParallel ComputationPipeline ParallelismAccuracyData Parallel ExecutionDeep Learning On GpusModel ParallelismMulti-gpu ArchitecturesParallel Deep LearningParallelizationPipeline Partitioning
The training process of Deep Neural Network (DNN) is compute-intensive, often taking days to weeks to train a DNN model. Therefore, parallel execution of DNN training on GPUs is a widely adopted approach to speed up the process nowadays. Due to the implementation simplicity, data parallelism is currently the most commonly used parallelization method. Nonetheless, data parallelism suffers from excessive inter-GPU communication overhead due to frequent weight synchronization among GPUs. Another approach is pipelined model parallelism, which partitions a DNN model among GPUs, and processes multiple mini-batches concurrently. This approach can significantly reduce inter-GPU communication cost compared to data parallelism. However, pipelined model parallelism faces the weight staleness issue; that is, gradients are computed with stale weights, leading to training instability and accuracy loss. In this paper, we present a pipelined model parallel execution method that enables high GPU utilization while maintaining robust training accuracy via a novel weight prediction technique, SpecTrain. Experimental results show that our proposal achieves up to 8.91x speedup compared to data parallelism on a 4-GPU platform while maintaining comparable model accuracy.
36
Sequence to Sequence Learning with Neural Networks
Ilya Sutskever, Oriol Vinyals, Quoc V. Le · arXiv (Cornell University) · 2014 · 13.4K citations · Full text
Computing, Computer Science, French +14
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke et al. · 2017 · 12.6K citations · Full text
Automatic differentiation in PyTorch
Adam Paszke, Sam Gross, Soumith Chintala et al. · 2017 · 11.1K citations