Data Analysis Group
The members of the Data Analysis Group perform research and development over a wide range of topics having to do with understanding and making effective use of large-scale data. Our customers include physicists, chemists, mathematicians, and others within the programs at LLNL as well as SciDAC (DOE Office of Science) and the greater computational science community.
Our goal is to enable scientists to concentrate on science by minimizing the effort required to explore and to understand the data generated by their simulations and experiments. This goal drives research efforts in large-scale visualization, data compression and streaming, and sophisticated analysis methods. To learn more about what we do, we invite you to look at some of the projects to which our group members contribute such as LOCAL, and Large-Scale Visualization. Our funding comes from a variety of sources including ASC, Laboratory Directed Research and Development, and the SciDAC Scientific Data Management Center.
Kathryn Mohror: performance measurement and analysis, extreme scale fault-tolerance, extreme scale tools, scalable storage and I/O
Ghaleb Abdulla: database technology, data access and integration
Sasha Ames: data management specific to file system metadata
Rushil Anirudh: computer vision, deep learning, machine learning on non-Euclidean domains, inverse problems in computer vision, and deep learning for X
Harsh Bhatia: topological data analysis and scientific visualization, information visualization, vector-valued spatial data, mathematical modeling and simulation
Peer-Timo Bremer: scientific visualization, multi-resolution algorithms, geometric modeling, topology data analysis
Jayaraman J. Thiagarajan: machine learning, data analysis and visualization, image understanding, computer vision and signal processing
Ming Jiang: scientific visualization, feature detection, multi-resolution algorithms, image processing
Bhavya Kailkhura: high-dimensional data analysis, adversarial ML/AI, non-convex optimization & decentralized control, robust statistics, coding and information theory
Hyojin Kim: computer vision, image understanding, machine learning, scalable and parallel computing with GPU
Dan Laney: scientific visualization, data compression, multi-resolution algorithms, simulated radiographic diagnostics for verification and validation
Peter Lindstrom: scientific visualization, geometric modeling, mesh simplification and compression, and multi-resolution algorithms