English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We present scalable hybrid-parallel algorithms for training large-scale 3Dconvolutional neural networks. Deep learning-based emerging scientificworkflows often require model training with large, high-dimensional samples,which can make training much more costly and even infeasible due to excessivememory usage. We solve these challenges by extensively applying hybridparallelism throughout the end-to-end training pipeline, including bothcomputations and I/O. Our hybrid-parallel algorithm extends the standard dataparallelism with spatial parallelism, which partitions a single sample in thespatial domain, realizing strong scaling beyond the mini-batch dimension with alarger aggregated memory capacity. We evaluate our proposed training algorithmswith two challenging 3D CNNs, CosmoFlow and 3D U-Net. Our comprehensiveperformance studies show that good weak and strong scaling can be achieved forboth networks using up 2K GPUs. More importantly, we enable training ofCosmoFlow with much larger samples than previously possible, realizing anorder-of-magnitude improvement in prediction accuracy.

The Case for Strong Scaling in Deep Learning: Training Large 3D CNNs with Hybrid Parallelism