New data standards created for AI models.
Budding bakers are frequently called upon to adapt award-winning recipes to suit different kitchen layouts. Someone could use an egg beater instead of a stand mixer to make award-winning chocolate chip cookies, for example.
Being able to replicate a recipe in different situations and with varying setups is essential for talented chefs and computer scientists, with the latter facing a similar problem of adapting and replicating their own “recipes” when trying to validate and working with new AI models.These models have applications in scientific fields ranging from climate analysis to brain research.
“When we talk about data, we have a working understanding of the digital assets we’re dealing with,” said Eliu Huerta, scientist and lead for translational AI at the US Department of Energy’s (DOE) Argonne National Laboratory. . “With an AI model, it’s a little less clear; are we talking about intelligently structured data, or is it computing, or software, or a mix?”
In a new study, Huerta and his colleagues defined a new set of standards for managing AI models. Adapted from recent research on automated data management, these standards are called FAIR, which stands for findable, accessible, interoperable and reusable.
“By making AI models FAIR, we no longer need to build every system from scratch every time,” said Argonne computing scientist Ben Blaiszik. “It becomes easier to reuse concepts from different groups, which helps create cross-pollination between teams.”
According to Huerta, the fact that many AI models are currently not FAIR poses a challenge to scientific discovery. “For many studies that have been done to date, it is difficult to access and replicate the AI models referenced in the literature,” he said. “By creating and sharing FAIR AI models , we can reduce duplication of effort and share best practices on how to use these models to enable great science.”
To meet the needs of a diverse community of users, Huerta and his colleagues combined a unique suite of data management and high-performance computing platforms to establish a FAIR protocol and quantify the “FAIR-ness” of models. of AI. Researchers combined FAIR data published in an online repository called Materials Data Facility, with FAIR AI models published in another online repository called Data and Learning Hub for Science, as well as AI and supercomputing resources at the Argonne Leadership Computing Facility (ALCF). This way, the researchers were able to create a computational framework that could help link various hardware and software together, creating AI models that could be run the same way on all platforms and yield reproducible results. The ALCF is a user facility of the DOE Office of Science.
Two keys to creating this framework are platforms called funcX and Globus, which allow researchers to access high-performance computing resources directly from their laptops. “FuncX and Globus can help transcend differences in hardware architectures,” said co-author Ian Foster, director of Argonne’s Data Science and Learning Division. “If someone is using one computing architecture and someone else is using another, we now have a way to speak a common AI language. This is an important part of making AI more interoperable.
In the study, the researchers used an example dataset from an AI model that used diffraction data from the Argonne Advanced Photon Source, also a DOE Office of Science user facility. To perform the calculations, the team used ALCF AI Testbed’s SambaNova system and NVIDIA GPUs (graphics processing units) from the Theta supercomputer.
“We’re excited to see the FAIR productivity benefits of sharing models and data to provide more researchers with access to high-performance computing resources,” said Marc Hamilton, NVIDIA vice president of architecture and engineering. solution engineering. “Together, we support the expanding universe of high-performance computing that combines experimental data and the operation of cutting-edge AI instruments to accelerate the pace of scientific discovery.”
“SambaNova is thrilled to partner with researchers at Argonne National Laboratory to pursue innovation at the interface of AI and emerging hardware architectures,” added Jennifer Glore, vice president of customer engineering at SambaNova Systems. “AI will play an important role in the future of scientific computing, and the development of FAIR principles for AI models along with new tools will enable researchers to enable large-scale autonomous discovery. We look forward to the continued collaboration and development at the ALCF AI testbed.”
A paper based on the study, “FAIR Principles for AI Models, with Practical Application for High-Energy Accelerated Diffraction Microscopy,” appeared in Nature science data on Nov 10, 2022.
In addition to Huerta, other study authors include Nikil Ravi of Argonne, Pranshu Chaturvedi, Zhengchun Liu, Ryan Chard, Aristana Scourtas, KJ Schmidt, Kyle Chard, Ben Blaiszik, and Ian Foster.
The research was funded by the DOE’s Office of Advanced Scientific Computing Research, the National Institutes of Standards and Technology, the National Science Foundation, and laboratory-directed research and development grants.
The Argonne Leadership Computing Facility provides supercomputing capabilities to the scientific and engineering community to advance fundamental discovery and understanding across a wide range of disciplines. Supported by the Advanced Scientific Computing Research (ASCR) program of the U.S. Department of Energy’s (DOE) Office of Science, the ALCF is one of two DOE advanced computing facilities dedicated to open science.
About the Advanced Photon Source
The U.S. Department of Energy Office of Science’s Advanced Photon Source (APS) at Argonne National Laboratory is one of the most productive x-ray light source facilities in the world. APS provides high-luminosity X-ray beams to a diverse community of researchers in materials science, chemistry, condensed matter physics, life and environmental sciences, and applied research. These X-rays are perfectly suited to the exploration of materials and biological structures; elementary distribution; chemical, magnetic, electronic states; and a wide range of technologically important engineering systems, from batteries to fuel injectors, all of which are the foundations of our country’s economic, technological and physical well-being. Each year, more than 5,000 researchers use APS to produce more than 2,000 publications detailing impactful discoveries and solving more vital biological protein structures than users of any other X-ray light source research facility. Scientists and APS engineers are innovating in technology that is central to advancing accelerator and light source operations. This includes insertion devices that produce the extremely bright X-rays that are prized by researchers, lenses that focus X-rays down to a few nanometers, instrumentation that maximizes how X-rays interact with samples studied and the software that gathers and manages the massive amount of data resulting from discovery research at APS.
This research utilized resources from the Advanced Photon Source, a United States DOE Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02- 06CH11357.
Argonne National Laboratory seeks solutions to pressing national problems in science and technology. The nation’s first national laboratory, Argonne conducts cutting-edge basic and applied scientific research in virtually every scientific discipline. Argonne researchers work closely with researchers from hundreds of companies, universities, and federal, state, and municipal agencies to help them solve their specific problems, advance American scientific leadership, and prepare the nation for a better future. With employees in more than 60 countries, Argonne is managed by UChicago Argonne, LLC for the US Department of Energy’s Office of Science.
U.S. Department of Energy Office of Science is the largest supporter of basic physical science research in the United States and strives to address some of the most pressing challenges of our time. For more information, visit https://energy.gov/science.
The title of the article
FAIR Principles for AI Models with Practical Application for High-Energy Accelerated Diffraction Microscopy
Publication date of articles
#Argonne #Scientists #Promote #FAIR #Standards #Managing #Artificial #Intelligence #Models