A team of researchers from the Department of Energy’s Oak Ridge National Laboratory has married artificial intelligence and high-performance computing to achieve a peak speed of 20 petaflops in the generation and training of deep learning networks on the laboratory’s Titan supercomputer.
Deep learning is a burgeoning field of artificial intelligence that uses networks modeled after the human brain to “learn” how to distinguish features and patterns in vast datasets. Such networks hold great promise in the realization of numerous technologies, from self-driving cars to intelligent robots.
Due to its ability to make sense of massive amounts of data, researchers across the scientific spectrum are eager to refine deep learning and apply it to some of today’s most challenging science problems. One such effort is ORNL’s Advances in Machine Learning to Improve Scientific Discovery at Exascale and Beyond (ASCEND) project, which aims to use deep learning to make sense of the massive datasets produced by the world’s most sophisticated scientific experiments, such as those located at ORNL.
Analysis of such datasets generally requires existing neural networks to be modified, or novel networks designed and then “trained” so that they know precisely what to look for and can produce valid results.
This is a time-consuming and difficult task, but one that an ORNL team led by Robert Patton and including Steven Young and Travis Johnston recently demonstrated can be dramatically expedited with a capable computing system such as ORNL’s Titan, the nation’s fastest supercomputer for science.
To efficiently design neural networks capable of tackling scientific datasets and expediting breakthroughs, Patton’s team developed two codes for evolving (MENNDL) and fine-tuning (RAvENNA) deep neural network architectures.
Both codes can generate and train as many as 18,600 neural networks simultaneously. Peak performance can be estimated by randomly sampling, and then carefully profiling, several hundred of these independently trained networks.
Both codes achieved a peak performance of 20 petaflops, or 20 thousand trillion calculations per second, on Titan (or just under half of Titan’s single precision total peak performance). In practical terms, that translates to training 40-50,000 networks per hour.
“The real measure of success in the deep learning community is time-to-solution,” said Johnston. “And with a machine like Titan we are able to train an unparalleled number of highly accurate networks.”
Titan is a Cray hybrid system, meaning that it uses both traditional CPUs and graphics processing units (GPUs) to tackle complex calculations for big science problems efficiently; the GPUs also happen to be the processor of choice for training deep learning networks.
The team’s work demonstrates that with the right high-performance computing system researchers can efficiently train large numbers of networks, which can then be used to help them tackle today’s increasingly data-heavy experiments and simulations.
This efficient design of deep neural networks will enable researchers to deploy highly accurate, custom-designed models, saving both time and money by freeing the scientist from the task of designing a network from the ground up.
And because the OLCF’s next leadership computing system, Summit, features a deep-learning friendly architecture with enhanced GPUs and complementary Tensor cores, the team is confident both codes will only get faster.
“Out of the box, without tuning to Summit’s unique architecture, we are expecting an increase in performance up to 50 times,” said Johnston.
With that sort of network training capability, Summit could be indispensable to researchers across the scientific spectrum looking to deep learning to help them tackle some of science’s most immense challenges.
Patton’s team is not waiting for the improved hardware to start tackling current scientific data challenges; they have already deployed their codes to assist domain scientists at the Department of Energy’s Fermilab in Batavia, Illinois.
Researchers at Fermilab used MENNDL to better understand how neutrinos interact with ordinary matter by producing a classification network to support their Main Injector Experiment for v-A (MINERvA), a neutrino scattering experiment. The task, known as vertex reconstruction, required a network to analyze images and precisely identify the location where neutrinos interact with one of many targets—a task akin to finding the aerial source of a starburst of fireworks.
In only 24 hours, MENNDL produced optimized networks that outperformed any previously handcrafted network—an achievement that could easily have taken scientists months to accomplish. To identify the high-performing network, MENNDL evaluated approximately 500,000 neural networks, training them on a data set consisting of 800,000 images of neutrino events, steadily using 18,000 of Titan’s nodes.
“You need something like MENNDL to explore this effectively infinite space of possible networks, but you want to do it efficiently,” Young said. “What Titan does is bring the time to solution down to something practical.”
And with Summit to come online this year, the future of deep learning in big science looks bright indeed.
The Latest on: Deep learning networks
- Neuron Bursts Can Mimic Famous AI Learning Strategyon October 18, 2021 at 7:55 am
A new model of learning centers on bursts of neural activity that act as teaching signals — approximating backpropagation, the algorithm behind learning in AI.
- Attention-Based Deep Learning Networks Could Improve Sonar Systemson October 16, 2021 at 4:12 pm
Researchers in China and the United States recently explored how an attention-based deep neural network (ABNN) could help improve sonar systems. The research was published in the Journal of the ...
- Researchers ‘Drop the Zeroes’ to Speed Deep Learningon October 15, 2021 at 3:00 am
Researchers at the King Abdullah University of Science and Technology (KAUST) are now proposing a method of accelerating distributed deep learning by dropping data blocks with zero values, which are ...
- These neural networks know what they’re doingon October 13, 2021 at 9:01 pm
MIT researchers have demonstrated that a special class of deep learning neural networks is able to learn the true cause-and-effect structure of a navigation task during training.
- Applying The Power Of Deep Learning To Cybersecurityon October 13, 2021 at 12:28 pm
There are lots of cybersecurity tools out there that promise to help organizations scale to manage the volume of threats and help separate the signal from the noise to identify security events that ...
- Attention-based deep neural network increases detection capability in sonar systemson October 12, 2021 at 9:33 am
In underwater acoustics, deep learning is gaining traction in improving sonar systems to detect ships and submarines in distress or in restricted waters. However, noise interference from the complex ...
- Cracking the Mystery of Deep Learningon October 11, 2021 at 10:00 pm
In the machine learning world, the sizes of artificial neural networks — and their outsize successes — are creating conceptual conundrums. When a network named AlexNet won an annual image ...
- What is Deep Learning?...and why we need it now.on October 11, 2021 at 1:25 pm
Introduction Deep Learning has become a hot topic in education circles recently, as Google now employs deep learning across their various platforms. However, the concept of deep learning is not new.
- A New Link to an Old Model Could Crack the Mystery of Deep Learningon October 11, 2021 at 8:36 am
To help them explain the shocking success of deep neural networks, researchers are turning to older but better-understood models of machine learning.
- A framework to enhance deep learning using first-spike timeson October 5, 2021 at 6:30 am
Researchers at Heidelberg University and University of Bern have recently devised a technique to achieve fast and energy-efficient computing using spiking neuromorphic substrates. This strategy, ...
via Google News and Bing News