CosmoFlow: Using Deep Learning to Learn the Universe at Scale

By Amrita Mathuriya, D. Bard, P. Mendygral, Lawrence Meadows, J. Arnemann, Lei Shao, Siyu He, Tuomas Kärnä, Diana Moise, S. Pennycook and 7 more. 2018

138 citations recorded by the source corpus.

Deep learning is a promising tool to determine the physical model that describes our universe. To handle the considerable computational cost of this problem, we present CosmoFlow: a highly scalable deep learning application built on top of the TensorFlow framework. CosmoFlow uses efficient implementations of 3D convolution and pooling primitives, together with improvements in threading for many element-wise operations, to improve training performance on Intel® Xeon Phi™ processors. We also utilize the Cray PE Machine Learning Plugin for efficient scaling to multiple nodes. We demonstrate fully synchronous data-parallel training on 8192 nodes of Cori with 77% parallel efficiency, achieving 3.5 Pflop/s sustained performance. To our knowledge, this is the first large-scale science application of the TensorFlow framework at supercomputer scale with fully-synchronous training. These enhancements enable us to process large 3D dark matter distribution and predict the cosmological parameters ΩsubM/sub, σsub8/sub and nsubs/sub with unprecedented accuracy.

Subfields: Artificial Intelligence, Computational Mechanics, Hardware and Architecture

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computational Mechanics

In Hardware and Architecture

Citations in the corpus

Learning to predict the cosmological structure formation (2018, background): “In particular, in astronomy and cosmology, a growing number of recent studies are using deep learning for a variety of tasks, ranging from analysis of cosmic microwave background (30–32), large-scale structure (33, 34), and gravitational lensing effects (35, 36) to classification of different light ”
Exascale Deep Learning for Climate Analytics (2018, background): “In recent years, the scientific community has begun to adopt deep learning methods and frameworks as tools for scientific analysis and discovery [13], [14], [15], [16], [17], [18].”
Etalumis: Bringing Probabilistic Programming to Scientific Simulators at Scale (2019, methodology): “To enable rapid prototyping, model evaluation, and making use of HPC capacity, scaling deep learning training to multiple computation units is highly desirable [38, 44, 45, 51, 52].”