Researchers extracted the relationship between strain and composition in a battery material by applying machine-learning methods to atomic-scale images. The work could lead to more durable batteries and also highlights the potential of integrating microscopy techniques with machine learning to gain insights into complex materials. Read more »
Machine-Learning Team Receives 2021 Halbach Award
This year’s Halbach Award for Innovative Instrumentation at the ALS went to a team of accelerator physicists and computer scientists who were able to use machine-learning techniques to solve a problem that has plagued third-generation light sources for a long time: fluctuations in beam size due to the motion of insertion devices. Read more »
Autonomous Data Acquisition for Scientific Discovery
Researchers at large scientific facilities such as the ALS have applied a robust machine-learning technique to automatically optimize data gathering for a variety of experimental techniques. The work promises to enable experiments with large, complex datasets to be run more quickly, efficiently, and with minimal human intervention. Read more »
X-Ray Experiments, Machine Learning Could Trim Years Off Battery R&D
Scanning transmission x-ray microscopy at the ALS’s COSMIC beamline contributed to a battery study that used an innovative approach to machine learning to speed up the learning curve about a process that shortens the life of fast-charging lithium batteries. It represents the first time this brand of “scientific machine learning” was applied to battery cycling. Read more »
ALS and Molecular Foundry Funded to Lead the Development of New Artificial Intelligence and Machine Learning Tools
DOE recently awarded a combined $8.55 million to two Berkeley Lab-led teams to build new tools that harness the power of artificial intelligence and machine learning. Synchrotron and nanoscience users will benefit from Alex Hexemer’s MLExchange and Andrew Minor’s 4D Camera Distillery, both multidisciplinary projects involving multiple national labs. Read more »
Machine Learning Helps Stabilize Synchrotron Light
Researchers showed that machine learning can predict noisy fluctuations in the size of beams generated by synchrotron light sources and correct them before they occur. The work solves a decades-old problem and will allow researchers to fully exploit the smaller beams made possible by recent advances in light source technology. Read more »
Machine Learning Enhances Light-Beam Performance at the Advanced Light Source
Researchers have successfully demonstrated how machine-learning tools can improve beam-size stability via adjustments that largely cancel out these fluctuations—reducing them from a level of a few percent down to 0.4 percent, with submicron precision. The demonstration shows that the technique could be viable for scientific light sources around the globe. Read more »
Scientists Use Machine Learning to Span Scales in Shale
Machine-learning techniques have been used to integrate fine- and large-scale infrared characterizations of shale—sedimentary rocks composed of minerals and organic matter. Understanding shale chemistry at both the nano and mesoscale is relevant to energy production, climate-change mitigation, and sustainable water and land use. Read more »