Machine Learning & Analytics Group

The Machine Learning & Analytics Group engages in research, development, and deployment of scalable, high performance, and data-intensive machine learning, analytics, and visualization technologies in collaboration with other researchers from diverse scientific disciplines. These methods and their application help to uncover new knowledge in scientific data from simulations, experiments and observations.

We accomplish this mission by

Focusing RD&D efforts at all stages of the data understanding pipeline.
Collaborating closely with science stakeholders to maximize science impact.
Tightly integrating and coordinating interaction between research, development, and deployment activities.

Validating deep learning inference during chest X-ray classification for COVID-19 screening (More information)

Topological Analysis and CinemaDB Enable Richer Exploration of In Situ Visualization Results (More information)

Programming Model Bake-off Helps Inform Data Analysis Directions in Post-Moore’s Era (More information)

New Image Reconstruction Framework Using Sparse Matrix-Vector Multiplication Outperforms State-Of-The-Art Methods (More information)

Accelerating Optimization of a Probabilistic Graphical Model through Data-Parallel Primitives (More information)

Nested Tracking Graphs Show Relationship of Features in Simulations Over Time (More information)

Accelerating Graph Analytics through Data-Parallel Primitives (More information)

MSMcam: Materials Segmentation & Metrics (More information)

BASTet: Berkeley Analysis and Storage Toolkit

NWB:N -- Beta of Neurodata Without Borders 2.0 Format and Software Released

BrainFormat: A Data Standardization Framework for Neuroscience Data

Petascale Study of Ion Foreshock Dynamics (More information)

Visible-Wavelength Polarized Light Emission with Small-Diameter InN Nanowires (More information)

OpenMSI: A Science Gateway to Sort Through Bio-Imaging's Big Datasets

Flying Through the Known Universe in 3D (More information)

CO<sub>2</sub> Sequestration and Storage: From Raw Micro-CT to Quality Measurements — CO₂ Sequestration and Storage: From Raw Micro-CT to Quality Measurements (More information)

Parallel Stream Surface Computation for Large Data Sets (More information)

Environmental Management: Hanford Site Nuclear Waste Clean-up (More information)

High-performance Computing for Computational Finance (More information)

Linking PointCloudXplore and Matlab: Making Advanced Analysis Easily Accessible (More information)

Query-Driven Analysis of Large Scale Time-dependent Data (More information)

High-throughput Characterization of Porous Materials (More information)

Performance Optimization and Auto-tuning (More information)

Parallel I/O, Analysis, and Visualization of a Trillion-Particle Simulation (More information)

Pattern Detection in Climate Data (More information)

Topological Cacti: Combining Structural and Quantitative Information (More information)

MetroMaps: Map-based Representations for Analyzing Optimization Solution Spaces (More information)

Efficient Parallel Extraction of Crack-free Isosurfaces from Adaptive Mesh Refinement Data (More information)

Streamline Integration using MPI-Hybrid Parallelism on a Large, Multi-Core Architecture (More information)

High Performance Visualization: Enabling Extreme Scale Scientific Insight (More information)

News

November 15, 2021

Wes Bethel and Gunther Weber are part of the Organizing Committee for the SC21 Workshop "In Situ Infrastructures for Extreme-scale Analysis and Visualization (ISAV'21)", now in its 6th year. More information.

November 14, 2021

Wes Bethel and Burlen Loring are part of an SC21 Tutorial on the subject of HPC In Situ Processing. More information.

August 30, 2021

Dani Ushizima and her collaborators at ALS receives the Halbach Award for the development of a machine-learning-based application to stabilize the transverse beam size and enhance the photon-beam performance at ALS; read more here.

August 18, 2021

Talita Perciano is being recognized for providing exceptional mentorship to students participating in various internships provided by Berkeley Lab; read more here.

August 2, 2021

Dani Ushizima and the Center for Recognition and Inspection of Cells (CRIC) have taken a major step toward applying machine learning techniques to women's health: their newly published database of images of cervical-vaginal cytology offers innovative opportunities for the use of machine learning for biomedical purposes; read more here.

February 12, 2021

Dani Ushizima named one of the 25 Latina in Science for her role in developing algorithms and software for automated cell analysis; read more here.

January 19, 2021

We are now the Machine Learning & Analytics Group, a name change that reflects our increasing focus on AI/ML technologies and their application to DOE mission science problems.

January 4, 2021

After 20 years of service, Wes Bethel will be stepping down as Group Leader. We have begun an international search for a new Group Leader; more information is available at this job posting.

April 24, 2020

Aditi Krishnapriyan joins the Data Analytics and Visualization Group.

Machine Learning & Analytics Group

News

... see all News