An innovative breakthrough in deep learning using one of the world’s most powerful supercomputers is enabling the study of galaxy formation, with future implications for use cases in other fields such as healthcare, automotive manufacture and smart cities.
The project, conducted in Japan by the University of Tokyo, Tohoku University and Kobe University, leveraged the power of the Fugaku supercomputer. It had the aim of accelerating forecasts around the expansion of supernova shells for highly resolved galaxy simulations. The breakthrough was based around a new type of deep learning inference engine called SoftNeuro from AI-based image processing solution developer Morpho.
SoftNeuro supports major deep learning frameworks and performs faster processing in various edge-device environments. As a general-purpose inference engine, it can be used not only for image recognition but also for speech recognition and text analysis.
“Morpho has worked on optimization on SoCs for smart devices in the past, but through the 3D CNN acceleration of SoftNeuro on Fugaku, we were able to take on the new challenge,” explained Hitoshi Matsuo, Senior Researcher, CTO Office, Morpho. “I expect that the experience gained from this project will enable us to utilize our technology for optimization of 3D calculations and optimization on supercomputers in the future.”
With the growing need for faster and more efficient 3D simulations using deep learning, Morpho said will continue to further improve the technical capabilities of SoftNeuro and develop the technology on a global level for use in numerous applications.
If you are interested in advances in AI and other cutting-edge tech, then this could be the event for you:
Responses