ORNL-Designed Algorithm Leverages Titan to Create High-Performing Deep Neural Networks

November 29, 2017

Nov. 29, 2017 — Deep neural networks—a form of artificial intelligence—have demonstrated mastery of tasks once thought uniquely human. Their triumphs have ranged from identifying animals in images, to recognizing human speech, to winning complex strategy games, among other successes.

Now, researchers are eager to apply this computational technique—commonly referred to as deep learning—to some of science’s most persistent mysteries. But because scientific data often looks much different from the data used for animal photos and speech, developing the right artificial neural network can feel like an impossible guessing game for nonexperts. To expand the benefits of deep learning for science, researchers need new tools to build high-performing neural networks that don’t require specialized knowledge.

Using the Titan supercomputer, a research team led by Robert Patton of the US Department of Energy’s(DOE’s) Oak Ridge National Laboratory (ORNL) has developed an evolutionary algorithm capable of generating custom neural networks that match or exceed the performance of handcrafted artificial intelligence systems. Better yet, by leveraging the GPU computing power of the Cray XK7 Titan—the leadership-class machine managed by the Oak Ridge Leadership Computing Facility, a DOE Office of Science User Facility at ORNL—these auto-generated networks can be produced quickly, in a matter of hours as opposed to the months needed using conventional methods.

The research team’s algorithm, called MENNDL (Multinode Evolutionary Neural Networks for Deep Learning), is designed to evaluate, evolve, and optimize neural networks for unique datasets. Scaled across Titan’s 18,688 GPUs, MENNDL can test and train thousands of potential networks for a science problem simultaneously, eliminating poor performers and averaging high performers until an optimal network emerges. The process eliminates much of the time-intensive, trial-and-error tuning traditionally required of machine learning experts.

“There’s no clear set of instructions scientists can follow to tweak networks to work for their problem,” said research scientist Steven Young, a member of ORNL’s Nature Inspired Machine Learning team. “With MENNDL, they no longer have to worry about designing a network. Instead, the algorithm can quickly do that for them, while they focus on their data and ensuring the problem is well-posed.”

Pinning down parameters

Inspired by the brain’s web of neurons, deep neural networks are a relatively old concept in neuroscience and computing, first popularized by two University of Chicago researchers in the 1940s. But because of limits in computing power, it wasn’t until recently that researchers had success in training machines to independently interpret data.

Today’s neural networks can consist of thousands or millions of simple computational units—the “neurons”—arranged in stacked layers, like the rows of figures spaced across a foosball table. During one common form of training, a network is assigned a task (e.g., to find photos with cats) and fed a set of labeled data (e.g., photos of cats and photos without cats). As the network pushes the data through each successive layer, it makes correlations between visual patterns and predefined labels, assigning values to specific features (e.g., whiskers and paws). These values contribute to the weights that define the network’s model parameters. During training, the weights are continually adjusted until the final output matches the targeted goal. Once the network learns to perform from training data, it can then be tested against unlabeled data.

Although many parameters of a neural network are determined during the training process, initial model configurations must be set manually. These starting points, known as hyperparameters, include variables like the order, type, and number of layers in a network.

Finding the optimal set of hyperparameters can be the key to efficiently applying deep learning to an unusual dataset. “You have to experimentally adjust these parameters because there’s no book you can look in and say, ‘These are exactly what your hyperparameters should be,’” Young said. “What we did is use this evolutionary algorithm on Titan to find the best hyperparameters for varying types of datasets.”

Unlocking that potential, however, required some creative software engineering by Patton’s team. MENNDL homes in on a neural network’s optimal hyperparameters by assigning a neural network to each Titan node. The team designed MENNDL to use a deep learning framework called Caffe to carry out the computation, relying on the parallel computing Message Passing Interface standard to divide and distribute data among nodes. As Titan works through individual networks, new data is fed to the system’s nodes asynchronously, meaning once a node completes a task, it’s quickly assigned a new task independent of the other nodes’ status. This ensures that the 27-petaflop Titan stays busy combing through possible configurations.

“Designing the algorithm to really work at that scale was one of the challenges,” Young said. “To really leverage the machine, we set up MENNDL to generate a queue of individual networks to send to the nodes for evaluation as soon as computing power becomes available.”

To demonstrate MENNDL’s versatility, the team applied the algorithm to several datasets, training networks to identify sub-cellular structures for medical research, classify satellite images with clouds, and categorize high-energy physics data. The results matched or exceeded the performance of networks designed by experts.

Networking neutrinos

One science domain in which MENNDL is already proving its value is neutrino physics. Neutrinos, ghost-like particles that pass through your body at a rate of trillions per second, could play a major role in explaining the formation of the early universe and the nature of matter—if only scientists knew more about them.

Large detectors at DOE’s Fermi National Accelerator Laboratory (Fermilab) use high-intensity beams to study elusive neutrino reactions with ordinary matter. The devices capture a large sample of neutrino interactions that can be transformed into basic images through a process called “reconstruction.” Like a slow-motion replay at a sporting event, these reconstructions can help physicists better understand neutrino behavior.

“They almost look like a picture of the interaction,” said Gabriel Perdue, an associate scientist at Fermilab.

Perdue leads an effort to integrate neural networks into the classification and analysis of detector data. The work could improve the efficiency of some measurements, help physicists understand how certain they can be about their analyses, and lead to new avenues of inquiry.

Teaming up with Patton’s team under a 2016 Director’s Discretionary application on Titan, Fermilab researchers produced a competitive classification network in support of a neutrino scattering experiment called MINERvA (Main Injector Experiment for v-A). The task, known as vertex reconstruction, required a network to analyze images and precisely identify the location where neutrinos interact with the detector—a challenge for events that produce many particles.

In only 24 hours, MENNDL produced optimized networks that outperformed handcrafted networks—an achievement that would have taken months for Fermilab researchers. To identify the high-performing network, MENNDL evaluated approximately 500,000 neural networks. The training data consisted of 800,000 images of neutrino events, steadily processed on 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,” Perdue said. “What Titan does is bring the time to solution down to something practical.”

Having recently been awarded another allocation under the Advanced Scientific Computing Research Leadership Computing Challenge program, Perdue’s team is building off its deep learning success by applying MENDDL to additional high-energy physics datasets to generate optimized algorithms. In addition to improved physics measurements, the results could provide insight into how and why machines learn.

“We’re just getting started,” Perdue said. “I think we’ll learn really interesting things about how deep learning works, and we’ll also have better networks to do our physics. The reason we’re going through all this work is because we’re getting better performance, and there’s real potential to get more.”

AI meets exascale

When Titan debuted 5 years ago, its GPU-accelerated architecture boosted traditional modeling and simulation to new levels of detail. Since then, GPUs, which excel at carrying out hundreds of calculations simultaneously, have become the go-to processor for deep learning. That fortuitous development made Titan a powerful tool for exploring artificial intelligence at supercomputer scales.

With the OLCF’s next leadership-class system, Summit, set to come online in 2018, deep learning researchers expect to take this blossoming technology even further. Summit builds on the GPU revolution pioneered by Titan and is expected to deliver more than five times the performance of its predecessor. The IBM system will contain more than 27,000 of Nvidia’s newest Volta GPUs in addition to more than 9,000 IBM Power9 CPUs. Furthermore, because deep learning requires less mathematical precision than other types of scientific computing, Summit could potentially deliver exascale-level performance for deep learning problems—the equivalent of a billion billion calculations per second.

“That means we’ll be able to evaluate larger networks much faster and evolve many more generations of networks in less time,” Young said.

In addition to preparing for new hardware, Patton’s team continues to develop MENNDL and explore other types of experimental techniques, including neuromorphic computing, another biologically inspired computing concept.

“One thing we’re looking at going forward is evolving deep learning networks from stacked layers to graphs of layers that can split and then merge later,” Young said. “These networks with branches excel at analyzing things at multiple scales, such as a closeup photograph in comparison to a wide-angle shot. When you have 20,000 GPUs available, you can actually start to think about a problem like that.”


Source: ORNL

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

A Beginner’s Guide to the ASC19 Finals

April 22, 2019

Three thousand watts. That's how much power the competitors in the 2019 ASC Student Supercomputer Challenge here in Dalian, China, have to work with. Everybody would like more juice to run compute-intensive HPC simulatio Read more…

By Alex Woodie

Is Data Science the Fourth Pillar of the Scientific Method?

April 18, 2019

Nvidia CEO Jensen Huang revived a decade-old debate last month when he said that modern data science (AI plus HPC) has become the fourth pillar of the scientific method. While some disagree with the notion that statistic Read more…

By Alex Woodie

At ASF 2019: The Virtuous Circle of Big Data, AI and HPC

April 18, 2019

We've entered a new phase in IT -- in the world, really -- where the combination of big data, artificial intelligence, and high performance computing is pushing the bounds of what's possible in business and science, in w Read more…

By Alex Woodie with Doug Black and Tiffany Trader

HPE Extreme Performance Solutions

HPE and Intel® Omni-Path Architecture: How to Power a Cloud

Learn how HPE and Intel® Omni-Path Architecture provide critical infrastructure for leading Nordic HPC provider’s HPCFLOW cloud service.

powercloud_blog.jpgFor decades, HPE has been at the forefront of high-performance computing, and we’ve powered some of the fastest and most robust supercomputers in the world. Read more…

IBM Accelerated Insights

Bridging HPC and Cloud Native Development with Kubernetes

The HPC community has historically developed its own specialized software stack including schedulers, filesystems, developer tools, container technologies tuned for performance and large-scale on-premises deployments. Read more…

Google Open Sources TensorFlow Version of MorphNet DL Tool

April 18, 2019

Designing optimum deep neural networks remains a non-trivial exercise. “Given the large search space of possible architectures, designing a network from scratch for your specific application can be prohibitively expens Read more…

By John Russell

A Beginner’s Guide to the ASC19 Finals

April 22, 2019

Three thousand watts. That's how much power the competitors in the 2019 ASC Student Supercomputer Challenge here in Dalian, China, have to work with. Everybody Read more…

By Alex Woodie

At ASF 2019: The Virtuous Circle of Big Data, AI and HPC

April 18, 2019

We've entered a new phase in IT -- in the world, really -- where the combination of big data, artificial intelligence, and high performance computing is pushing Read more…

By Alex Woodie with Doug Black and Tiffany Trader

Interview with 2019 Person to Watch Michela Taufer

April 18, 2019

Today, as part of our ongoing HPCwire People to Watch focus series, we are highlighting our interview with 2019 Person to Watch Michela Taufer. Michela -- the Read more…

By HPCwire Editorial Team

Intel Gold U-Series SKUs Reveal Single Socket Intentions

April 18, 2019

Intel plans to jump into the single socket market with a portion of its just announced Cascade Lake microprocessor line according to one media report. This isn Read more…

By John Russell

BSC Researchers Shrink Floating Point Formats to Accelerate Deep Neural Network Training

April 15, 2019

Sometimes calculating solutions as precisely as a computer can wastes more CPU resources than is necessary. A case in point is with deep learning. In early stag Read more…

By Ken Strandberg

Intel Extends FPGA Ecosystem with 10nm Agilex

April 11, 2019

The insatiable appetite for higher throughput and lower latency – particularly where edge analytics and AI, network functions, or for a range of datacenter ac Read more…

By Doug Black

Nvidia Doubles Down on Medical AI

April 9, 2019

Nvidia is collaborating with medical groups to push GPU-powered AI tools into clinical settings, including radiology and drug discovery. The GPU leader said Monday it will collaborate with the American College of Radiology (ACR) to provide clinicians with its Clara AI tool kit. The partnership would allow radiologists to leverage AI techniques for diagnostic imaging using their own clinical data. Read more…

By George Leopold

Digging into MLPerf Benchmark Suite to Inform AI Infrastructure Decisions

April 9, 2019

With machine learning and deep learning storming into the datacenter, the new challenge is optimizing infrastructure choices to support diverse ML and DL workfl Read more…

By John Russell

The Case Against ‘The Case Against Quantum Computing’

January 9, 2019

It’s not easy to be a physicist. Richard Feynman (basically the Jimi Hendrix of physicists) once said: “The first principle is that you must not fool yourse Read more…

By Ben Criger

Why Nvidia Bought Mellanox: ‘Future Datacenters Will Be…Like High Performance Computers’

March 14, 2019

“Future datacenters of all kinds will be built like high performance computers,” said Nvidia CEO Jensen Huang during a phone briefing on Monday after Nvidia revealed scooping up the high performance networking company Mellanox for $6.9 billion. Read more…

By Tiffany Trader

ClusterVision in Bankruptcy, Fate Uncertain

February 13, 2019

ClusterVision, European HPC specialists that have built and installed over 20 Top500-ranked systems in their nearly 17-year history, appear to be in the midst o Read more…

By Tiffany Trader

Intel Reportedly in $6B Bid for Mellanox

January 30, 2019

The latest rumors and reports around an acquisition of Mellanox focus on Intel, which has reportedly offered a $6 billion bid for the high performance interconn Read more…

By Doug Black

It’s Official: Aurora on Track to Be First US Exascale Computer in 2021

March 18, 2019

The U.S. Department of Energy along with Intel and Cray confirmed today that an Intel/Cray supercomputer, "Aurora," capable of sustained performance of one exaf Read more…

By Tiffany Trader

Looking for Light Reading? NSF-backed ‘Comic Books’ Tackle Quantum Computing

January 28, 2019

Still baffled by quantum computing? How about turning to comic books (graphic novels for the well-read among you) for some clarity and a little humor on QC. The Read more…

By John Russell

IBM Quantum Update: Q System One Launch, New Collaborators, and QC Center Plans

January 10, 2019

IBM made three significant quantum computing announcements at CES this week. One was introduction of IBM Q System One; it’s really the integration of IBM’s Read more…

By John Russell

Deep500: ETH Researchers Introduce New Deep Learning Benchmark for HPC

February 5, 2019

ETH researchers have developed a new deep learning benchmarking environment – Deep500 – they say is “the first distributed and reproducible benchmarking s Read more…

By John Russell

Leading Solution Providers

SC 18 Virtual Booth Video Tour

Advania @ SC18 AMD @ SC18
ASRock Rack @ SC18
DDN Storage @ SC18
HPE @ SC18
IBM @ SC18
Lenovo @ SC18 Mellanox Technologies @ SC18
NVIDIA @ SC18
One Stop Systems @ SC18
Oracle @ SC18 Panasas @ SC18
Supermicro @ SC18 SUSE @ SC18 TYAN @ SC18
Verne Global @ SC18

IBM Bets $2B Seeking 1000X AI Hardware Performance Boost

February 7, 2019

For now, AI systems are mostly machine learning-based and “narrow” – powerful as they are by today's standards, they're limited to performing a few, narro Read more…

By Doug Black

The Deep500 – Researchers Tackle an HPC Benchmark for Deep Learning

January 7, 2019

How do you know if an HPC system, particularly a larger-scale system, is well-suited for deep learning workloads? Today, that’s not an easy question to answer Read more…

By John Russell

Arm Unveils Neoverse N1 Platform with up to 128-Cores

February 20, 2019

Following on its Neoverse roadmap announcement last October, Arm today revealed its next-gen Neoverse microarchitecture with compute and throughput-optimized si Read more…

By Tiffany Trader

Intel Launches Cascade Lake Xeons with Up to 56 Cores

April 2, 2019

At Intel's Data-Centric Innovation Day in San Francisco (April 2), the company unveiled its second-generation Xeon Scalable (Cascade Lake) family and debuted it Read more…

By Tiffany Trader

France to Deploy AI-Focused Supercomputer: Jean Zay

January 22, 2019

HPE announced today that it won the contract to build a supercomputer that will drive France’s AI and HPC efforts. The computer will be part of GENCI, the Fre Read more…

By Tiffany Trader

Oil and Gas Supercloud Clears Out Remaining Knights Landing Inventory: All 38,000 Wafers

March 13, 2019

The McCloud HPC service being built by Australia’s DownUnder GeoSolutions (DUG) outside Houston is set to become the largest oil and gas cloud in the world th Read more…

By Tiffany Trader

Intel Extends FPGA Ecosystem with 10nm Agilex

April 11, 2019

The insatiable appetite for higher throughput and lower latency – particularly where edge analytics and AI, network functions, or for a range of datacenter ac Read more…

By Doug Black

UC Berkeley Paper Heralds Rise of Serverless Computing in the Cloud – Do You Agree?

February 13, 2019

Almost exactly ten years to the day from publishing of their widely-read, seminal paper on cloud computing, UC Berkeley researchers have issued another ambitious examination of cloud computing - Cloud Programming Simplified: A Berkeley View on Serverless Computing. The new work heralds the rise of ‘serverless computing’ as the next dominant phase of cloud computing. Read more…

By John Russell

Do NOT follow this link or you will be banned from the site!
Share This