Skip to navigation Skip to content
Careers | Phone Book | A - Z Index

Home

Artistic rendering suggestive of a neural network

Improving Training for Scientific Machine Learning

Neural networks can be difficult to train and employ on large-scale science problems. That's why Berkeley Lab researchers are devising new training methods tailored to scientific machine learning.

WarpX visualization

WarpX Goes Exascale

The ECP-funded WarpX Project has spent the last six years creating a novel, highly parallel, and highly optimized single-source simulation code for modeling plasma-based particle colliders on cutting-edge exascale supercomputers.

EQSIM visualization

EQSIM's Sophisticated Earthquake Simulations

A collaboration involving scientists and computing resources from Berkeley Lab will publicly release its most accurate earthquake simulations to date.

Tomography Data Set

Machine Learning Helps Process Large Bioimaging Datasets

Berkeley Lab scientists used several machine learning techniques in a pipeline to segment and identify cryo-ET cell membrane structures.

Huntress Digital Graphic

HUNTRESS: Expanding an Understanding of Tumor Progression

The latest developments in computational oncology are giving medical researchers a glimpse into a future where they’ll be able to understand tumor progression via supercomputers and advanced mathematical algorithms.

AQT Cryo Refrigerator for Quantum

Breakthrough in Quantum Universal Gate Sets

Berkeley Lab's Advanced Quantum Testbed team demonstrates a three-qubit native quantum gate with high fidelity

News

premise sparse cropped

New Math Methods and Perlmutter HPC Combine to Deliver Record-Breaking ML Algorithm

March 13, 2023

Using the Perlmutter supercomputer at the National Energy Research Scientific Computing Center (NERSC), researchers at Lawrence Berkeley National Laboratory (Berkeley Lab) have devised a new mathematical method for analyzing extremely large datasets – and, in the process, demonstrated proof of principle on a record-breaking dataset of more than five million points.

Neural Network Stock Image Cropped

Improving Training for Scientific Machine Learning

March 3, 2023

In the world of scientific machine learning (SciML), scientists are beginning to embrace the use of neural networks as a way to accelerate simulations. At the heart of deep learning algorithms, neural networks provide a mechanism to encode complex dependency structures, using many connected node layers to transform data into learned features to be used for a wide range of scientific tasks.


Earthquake artistic graphic

The Most Advanced Bay Area Earthquake Simulations Will be Publicly Available

February 10, 2023

A collaboration involving scientists and computing resources from Berkeley Lab and the simulation software EQSIM is releasing the most accurate and detailed earthquake simulations to date, which will initially capture earthquake motions across the San Francisco Bay Area and later expand to other regions.

WarpX longitudinal electric field

WarpX Code Shines at the Exascale Level

February 2, 2023

The WarpX project has spent the last six years creating a novel, highly parallel, and highly optimized single-source simulation code for modeling plasma-based particle colliders on cutting-edge exascale supercomputers, with broad importance for other accelerators and related problems.


BrainMap

Berkeley Lab’s Ushizima Honored with PMWC Pioneer Award

January 25, 2023

Berkeley Lab’s Daniela Ushizima was recognized with PMWC Pioneer Award for constructing “a new and reliable technique for diagnosing Alzheimer’s disease and measuring the efficacy of experimental treatments.”

Digital visualization of a CryoET machine learning simulation

Berkeley Lab Scientists Create Machine Learning Pipeline for Interpreting Large Tomography Datasets

January 25, 2023

A group of Berkeley Lab scientists has developed and tested several machine learning techniques organized in a learning pipeline to improve the interpretation of increasingly large cryo-ET datasets.