The Argonne Leadership Computing Facility (ALCF) recently awarded computing time and resources to three new projects and one renewed project for 2021-2022, through its ALCF Data Science Programme (ADSP). The ALCF is a U.S. Department of Energy (DOE) Office of Science User Facility at DOE’s Argonne National Laboratory.
The ADSP enables big data and artificial intelligence (AI) research that requires DOE’s leadership-class computing resources. The forward-looking allocation programme is designed to explore and improve computational methods for data-driven discoveries across scientific disciplines. It also focuses on scaling the underlying data science technologies to fully utilise DOE supercomputers.
The new projects extract science from a range of unique data sources. The projects aim to accelerate autonomous molecular design, data analysis in neutrino experiments, and sky survey discovery. The project selected for renewal will address challenges in fast, high-resolution X-ray imaging at the Advanced Photon Source (APS).
Each project will employ leadership-class systems and infrastructure to develop and advance data science techniques, with novel approaches to machine learning, deep learning, and other cutting-edge AI methods.
ALCF research scientist states that this year’s ADSP awards advance the use of AI on ALCF systems beyond standalone networks to multi-network workflows integrated into scientific analysis chains. In addition, unsupervised techniques are targeting their upcoming system Polaris, which is ideal for deep learning applications and will serve as a testbed for our future exascale supercomputer, Aurora.
The first new ASDP project is “Autonomous Molecular Design for Redox Flow Batteries”. The goal of this project is to build an autonomous AI application for supercomputers that can select and perform the simulation and machine learning tasks needed to identify better-performing molecules. Achieving this goal will require scaling individual tasks, such as computing material properties and training AI models, and then combining them into a cohesive application that will remove humans from the materials design process.
The second new ASDP project is “Machine Learning for Data Reconstruction to Accelerate Physics Discoveries in Accelerator-Based Neutrino Oscillation Experiments”. The optimization of a traditional data reconstruction pipeline in these experiments is done “by hand,” and can take months to years. The team’s goal is to reduce this process to hours using the ALCF’s upcoming Polaris system. This effort will accelerate the analysis pipeline, perhaps even enabling a full physics analysis online, allowing for more frequent and deeper inference of physics insights from experimental data.
The third new ASDP project is “Learning Optimal Image Representations for Current and Future Sky Surveys”. The team’s work aims to serve the broader community by accelerating sky survey discoveries following the release of image representations, trained models, and software. Researchers will be able to simply download the low-dimensional representations of galaxies to perform scientific analysis, or use the team’s pre-trained model and quickly fine-tune it to carry out a specific task.
Finally, the renewed ADSP project is “Dynamic Compressed Sensing for Real-Time Tomographic Reconstruction”. The team will They will experimentally demonstrate the reconstruction workflow and methods on commercial scanning transmission electron microscopes and the ptychographic tomography instruments at the APS. By integrating their tool with an open-source 3D visualization and tomography software package, the team’s techniques will be accessible to a wide range of researchers and enable new material characterizations in academia and industry.
The U.S. Department of Energy’s Argonne National Laboratory has also conducted research to map the pandemic’s impact. As reported by OpenGov Asia, the laboratory released interactive indices, analyses, and maps that provide a detailed understanding of the socio-economic effects of the novel coronavirus outbreak. The public now has access to a series of data and analysis resources designed to support and inform long-term COVID-19 recovery efforts across the U.S.
The data and analysis are helping to guide federal recovery efforts, from informing federal engagement efforts with affected communities to helping target delivery of aid. The laboratory developed these resources to help federal agencies understand where impacts are most acute (down to the county level), and which demographic groups and facets of the economy may require recovery support, such as employment, housing stability, public sector services
