Data Science in the News

CASC team wins best paper at visualization symposium

A research team from LLNL’s Center for Applied Scientific Computing won Best Paper at the 15th IEEE Pacific Visualization Symposium (PacificVis), which was held virtually on April 11–14. Computer scientists Harsh Bhatia, Peer-Timo Bremer, and Peter Lindstrom collaborated with University of Utah colleagues Duong Hoang, Nate Morrical, and Valerio Pascucci on “AMM: Adaptive Multilinear Meshes.”...

Building confidence in materials modeling using statistics

LLNL statisticians, computational modelers, and materials scientists have been developing a statistical framework for researchers to better assess the relationship between model uncertainties and experimental data. The Livermore-developed statistical framework is intended to assess sources of uncertainty in strength model input, recommend new experiments to reduce those sources of uncertainty...

Summer scholar develops data-driven approaches to key NIF diagnostics

Su-Ann Chong's summer project, “A Data-Driven Approach Towards NIF Neutron Time-of-Flight Diagnostics Using Machine Learning and Bayesian Inference,” is aimed at presenting a different take on nToF diagnostics. Neutron time-of-flight diagnostics are an essential tool to diagnose the implosion dynamics of inertial confinement fusion experiments at NIF, the world’s largest and most energetic...

Data Science Challenge welcomes UC Riverside

Together with LLNL’s Center for Applied Scientific Computing (CASC), the DSI welcomed a new academic partner to the 2021 Data Science Challenge (DSC) internship program: the University of California (UC) Riverside campus. The intensive program has run for three years with UC Merced, and it tasks undergraduate and graduate students with addressing a real-world scientific problem using data...

Visualization software stands the test of time

In the decades since LLNL’s founding, the technology used in pursuit of the Laboratory’s national security mission has changed over time. For example, studying scientific phenomena and predicting their behaviors require increasingly robust, high-resolution simulations. These crucial tasks compound the demands on high-performance computing hardware and software, which must continually be...

Laser-driven ion acceleration with deep learning

While advances in machine learning over the past decade have made significant impacts in applications such as image classification, natural language processing and pattern recognition, scientific endeavors have only just begun to leverage this technology. This is most notable in processing large quantities of data from experiments. Research conducted at LLNL is the first to apply neural...

The data-driven future of extreme physics

By applying modern machine learning and data science methods to “extreme” plasma physics, researchers can gain insight into our universe and find clues about creating a limitless amount of energy. In a recent perspective published in Nature, LLNL scientists and international collaborators outline key challenges and future directions in using machine learning and other data-driven techniques...

A winning strategy for deep neural networks

LLNL continues to make an impact at top machine learning conferences, even as much of the research staff works remotely during the COVID-19 pandemic. Postdoctoral researcher James Diffenderfer and computer scientist Bhavya Kailkhura, both from LLNL’s Center for Applied Scientific Computing, are co-authors on a paper—“Multi-Prize Lottery Ticket Hypothesis: Finding Accurate Binary Neural...

Winter hackathon highlights data science talks and tutorial

The Data Science Institute (DSI) sponsored LLNL’s 27th hackathon on February 11–12. Held four times a year, these seasonal events bring the computing community together for a 24-hour period where anything goes: Participants can focus on special projects, learn new programming languages, develop skills, dig into challenging tasks, and more. The winter hackathon was the DSI’s second such...

Novel deep learning framework for symbolic regression

LLNL computer scientists have developed a new framework and an accompanying visualization tool that leverages deep reinforcement learning for symbolic regression problems, outperforming baseline methods on benchmark problems. The paper was recently accepted as an oral presentation at the International Conference on Learning Representations (ICLR 2021), one of the top machine learning...

CASC research in machine learning robustness debuts at AAAI conference

LLNL’s Center for Applied Scientific Computing (CASC) has steadily grown its reputation in the artificial intelligence (AI)/machine learning (ML) community—a trend continued by three papers accepted at the 35th AAAI Conference on Artificial Intelligence, held virtually on February 2–9, 2021. Computer scientists Jayaraman Thiagarajan, Rushil Anirudh, Bhavya Kailkhura, and Peer-Timo Bremer led...

LLNL physicist wins Young Former Student award

Texas A&M University’s Department of Nuclear Engineering on December 10 announced it has honored LLNL physicist Kelli Humbird with its 2020-21 Young Former Student award for her work at LLNL in combining machine learning with inertial confinement fusion (ICF) research. Humbird graduated from Texas A&M with a PhD in nuclear engineering in 2019. Since joining the Laboratory as an intern in 2016...

NeurIPS papers aim to improve understanding and robustness of machine learning algorithms

The 34th Conference on Neural Information Processing Systems (NeurIPS) is featuring two papers advancing the reliability of deep learning for mission-critical applications at LLNL. The most prestigious machine learning conference in the world, NeurIPS began virtually on Dec. 6. The first paper describes a framework for understanding the effect of properties of training data on the...

DOE announces five new energy projects at LLNL

The DOE today announced two rounds of awards for the High Performance Computing for Energy Innovation Program HPC4EI), including five projects at LLNL. HPC4EI connects industry with the computational resources and expertise of the DOE national laboratories to solve challenges in manufacturing, accelerate discovery and adoption of new materials and improve energy efficiency. The awards were...

Machine learning speeds up and enhances physics calculations

Interpreting data from NIF’s cutting-edge high energy density science experiments relies on physics calculations that are so complex they can challenge LLNL supercomputers, which stand among the best in the world. A collaboration between LLNL and French researchers found a novel way to incorporate machine learning and neural networks to significantly speed up inertial confinement fusion...

LLNL papers accepted into prestigious conference

Two papers featuring LLNL scientists were accepted in the 2020 International Conference on Machine Learning (ICML), one of the world’s premier conferences of its kind. Read more at LLNL News.

AI identifies change in microstructure in aging materials

LLNL scientists have taken a step forward in the design of future materials with improved performance by analyzing its microstructure using AI. The work recently appeared online in the journal Computational Materials Science. Read more at LLNL News.

Deep learning may provide solution for efficient charging, driving of autonomous electric vehicles

LLNL computer scientists and software engineers have developed a deep learning-based strategy to maximize electric vehicle (EV) ride-sharing services while reducing carbon emissions and the impact to the electrical grid, emphasizing autonomous EVs capable of offering 24-hour service. Read more at LLNL News.

Department of Energy researchers share data management strategies at first-ever “Data Day”

It’s become something of a mantra of the digital age: Data is the new currency. Especially in science, where it’s hard to find a single project that doesn’t involve generating or consuming massive amounts of data. In light of the growing awareness of the critical importance of data management across the Department of Energy complex, more than 100 researchers from DOE national laboratories...

Successful simulation and visualization coupling proves the power of Sierra

As the first National Nuclear Security Administration (NNSA) production supercomputer backed by GPU- (graphics processing unit) accelerated architecture, Sierra’s acquisition required a fundamental shift in how scientists at Lawrence Livermore National Laboratory (LLNL) program their codes to take advantage of the GPUs. The majority of Sierra’s computational power—95 percent of its 125...