Washington, District of Columbia 2021-09-28 12:44:13 –
September 28, 2021
Science is in the midst of a flood of data. Experiments generate more information than researchers can process. However, new attempts centered on artificial intelligence will help scientists navigate this data-rich reality.
September 28, National Science Foundation Network Presentation A $ 15 million five-year grant to integrate AI tools into the scientific research and discovery process. The award will fund the Accelerated AI Algorithms of the Data-Driven Discovery Institute (or A3D3 Institute), a partnership of nine universities led by the University of Washington.
The A3D3 Institute provides scientists with a new paradigm shift AI tool for analyzing the types of large and complex datasets that are an increasingly common feature of research, from medical laboratories to particle colliders. Aims to accelerate the discovery pipeline.
“I am fortunate to be able to work with an extraordinary group of talented researchers and am excited to continue participating in solving some of the most basic problems of science and engineering. The ultimate goal of A3D3 is to build the institutional knowledge that is essential for real-time applications of AI in all scientific disciplines. ” Shih-Chieh Hsu, Associate Professor of Physics at the University of Washington and Director of the A3D3 Institute. “A3D3 provides scientists with new tools to deal with the upcoming flood of data through dedicated outreach activities.”
The A3D3 Institute, part of NSF’s Harnessing the Data Revolution program, is a collaboration between researchers at the University of Washington. University of Illinois at Urbana-Champaign. University of Duke; University of Massachusetts Institute of Technology; University of Minnesota, Twin Cities; University of California Institute of Technology; University of Padu; University of California, San Diego. University of Wisconsin-Madison.
In addition to Sue, other UW faculty members involved in the A3D3 Institute Scott Hawk, Professor of Electrical and Computer Engineering. Amy O’s Bone, Associate Professor of Bioengineering and Electrical and Computer Engineering.When Eli Shriserman, Associate Professor of Applied Mathematics and Electrical Engineering and Computer Engineering.
From detectors that search for gravitational waves to electrical sensors that monitor brain activity, research is passing scientists a larger set of data to analyze. Experiments are generating more data as researchers develop better tools, from sharper medical imaging techniques to more accurate sensors for particle physics experiments. CERN single experiment Large Hadron Collider, for example, Can be generated 1 petabyte of data per second (1 million gigabytes) from tens of millions of collisions. But as the size and complexity of datasets grows, to analyze the data and place the most relevant bits (or bytes) before the scientist’s eyes risk exceeding their current computing power. You will need the necessary algorithms.
A3D3 research focuses on developing AI-based algorithms that can perform real-time analysis of large datasets in three data-rich areas: multi-messenger astrophysics, high-energy particle physics, and neuroscience. increase.
“Advances in computing power with machine learning technology on high-performance computing platforms offer exciting new paths for scientific discoveries, while in fast, high-throughput data collection for scientific applications. Unique challenges are creating new demand for researchers. ” Hauck.
Multi-messenger astrophysics integrates observations of the universe from a variety of sources, including gravitational wave detectors, neutrino detectors, and telescopes, to suddenly and suddenly in space such as supernovae, astronomical collisions, and black hole binaries. Often identify and study violent events. A3D3 researchers quickly identify these events, and astronomers cross-correlate observations of the same event from different sources to build a more complete image of our empty transient event type. We will work on the development of AI algorithms that will help you to do so.
High-energy physics experiments, such as those studied by Sue on the Large Hadron Collider, could overturn the understanding of the universe by discovering new types of particles, such as candidate dark matter particles, and new basic forces. .. The A3D3 effort focuses on an AI fuel approach that detects unexpected anomalies in collision data and “reconstructs” the particles underlying 40 million collisions per second in high-energy experiments. These tools streamline the downstream analysis process and accelerate and simplify the discovery pipeline.
In neuroscience, A3D3’s efforts focus on understanding the complex neural networks in the human brain that govern motor function and process sensory information.
“Now we can measure more brains for longer periods of time. We need new tools to analyze these large datasets,” said a core staff scientist at the Washington National Primate Research Institute. Orsborn, who is also said. “Rapid analysis of data also enables new experiments and treatments that can intervene based on ongoing brain activity.”
Researchers need AI-based algorithms for real-time analysis of neural datasets, such as electrical recordings from embedded electrodes and extensive basic scientific research. Researchers at A3D3 will focus on developing these types of tools that help decipher the neural basis of behavior, such as basic motor function and response to stimuli.
“Importantly, A3D3 researchers will focus on developing scalable analytical tools that can adapt not only to today’s datasets, but also to the large and complex datasets expected in the coming decades. “Shlizerman said.
With the rapidly increasing amount of data generated by scientific research, the A3D3 Institute is also looking to the future. The institute seeks training and research opportunities for both graduate and undergraduate students, including students from backgrounds that are undervalued in the STEM community. These efforts ensure that the impact of A3D3 will spread and last beyond its immediate goals, Sue said.
Please contact Sue for more information. email@example.com..
tag: Amy O’s Bone •• College of Liberal Arts •• Faculty of Engineering •• Bioengineering department •• Department of Electrical and Computer Engineering •• Department of Physics •• Eli Shriserman •• School of medicine •• Scott Hawk •• Shih-Chieh Hsu
New NSF-funded institute to harness AI for accelerated discoveries in physics, astronomy and neuroscience Source link New NSF-funded institute to harness AI for accelerated discoveries in physics, astronomy and neuroscience