Machine Learning for Science Workshop (ML4Sci)


September 4, 2018 to September 6, 2018
8:50am to 11:30am
Lawrence Berkeley National Laboratory
Get Directions


This workshop will feature presentations on scientific ML applications in the lab – covering high energy physics, nuclear physics, cosmology, chemistry, biosciences, materials engineering, climate and high performance computing – and provide overviews of the state of the art in ML methodology and technology, as well as hands-on training for deploying ML applications on NERSC platforms.  

Machine Learning for Science Workshop (ML4Sci)
Date: September 4-6, 2018
Abstract Deadline: August 24, 2018
Location: LBNL Building 50 Auditorium

REGISTRATION is free.  Pre-registration is required. Regisration deadline: August 27, 2018

BIDS Senior Fellows Bin Yu , Philip StarkJohn Canny and Kristofer Bouchard are among the confirmed speakers.  The conference planning team includes BIDS Senior Fellows Dani Ushizima and Kristofer Bouchard



Bin Yu

Professor and Second Chair, Departments of Statistics and EECS, UC Berkeley

Bin Yu is Chancellor’s Distinguished Professor and Class of 1936 Second Chair in the Departments of Statistics and of Electrical Engineering & Computer Science, and Center for Computational Biology at the University of California, Berkeley. She is an Investigator with the Weill Neurohub, a collaboration of the University of California, Berkeley (UC Berkeley), the University of California, San Francisco (UCSF), and the University of Washington (the UW). She leads the Yu Group at Berkeley, which is engaged in interdisciplinary research with scientists from genomics, neuroscience, and medicine. In order to solve data problems in these domain areas, her group employs quantitative critical thinking and develops statistical and machine learning algorithms and theory. She has published more than 100 scientific papers in premier journals in statistics, machine learning, information theory, signal processing, remote sensing, neuroscience, genomics, and networks. She was a Guggenheim Fellow and President of Institute of Mathematical Statistics (IMS), and is a member of the U.S. National Academy of Sciences and fellow of the American Academy of Arts and Sciences. 

Philip B. Stark

Associate Dean, Mathematical and Physical Sciences, UC Berkeley

Philip B. Stark's research centers on inference (inverse) problems and uncertainty quantification, especially confidence procedures tailored for specific goals. Applications include causal inference, the U.S. Census, climate modeling, cosmology, earthquake prediction and seismic hazard analysis, election auditing, endangered species, epidemiology, evaluating and improving teaching and educational technology, food web models, health effects of sodium, the geomagnetic field, geriatric hearing loss, information retrieval, Internet content filters, litigation, resilient and sustainable food systems, risk assessment (including natural disasters and food safety), the seismic structure of Sun and Earth, spectroscopy, spectrum estimation, and uncertainty quantification for computational models of complex systems. Methods he has developed for auditing elections have been incorporated into laws in California, Colorado, and Rhode Island, Texas, Virginia, and Washington. Methods for data reduction and spectrum estimation that he has developed or co-developed are part of the Øersted geomagnetic satellite data pipeline and the Global Oscillations Network Group (GONG) helioseismic telescope network data pipeline. 

John Canny

Professor, Computer Science, UC Berkeley

John Canny is a professor of engineering in the Computer Science Department. He has made significant contributions in various areas of computer science and mathematics, including artificial intelligence, robotics, computer graphics, human-computer interaction, computer security, computational algebra, and computational geometry. As the author of “A Variational Approach to Edge Detection” and the creator of the widely used Canny edge detector, he was honored for seminal contributions in the areas of robotics and machine perception.

Daniela Ushizima

Staff Scientist, Applied Mathematics and Computational Research Division, Berkeley Lab

BIDS Faculty Affiliate Dani Ushizima is a Staff Scientist in the Machine Learning and Analytics Group in the Computational Research Division at Berkeley Lab, where she leads the Image Processing/Machine Vision team at CAMERA, and an Affiliate Faculty of the Bakar Computational Health Sciences Institute (BCHSI) at the University of California, San Francisco. She also leads the Center for Recognition and Inspection of Cells (CRIC), where her research focuses on imaging cancer cells for early-stage disease diagnosis. With 20 years of research and development experience in Computer Vision, Dani has focused primarily on quantitative microscopy and microstructure classification, from materials science to biomedical imaging.

Kristofer Bouchard

Staff Scientist & Acting Group Lead, Computational Biosciences, LBNL

Kristofer Bouchard is a Staff Scientist and the Acting Group Lead for Computational Biosciences at Lawrence Berkeley National Laboratory.  He is also the PI of the Neural Systems and Data Science Lab at LBNL/UCB, an interdisciplinary team that focuses on understanding how distributed neural circuits gives rise to coordinated behaviors and perception. They take a two-pronged approach to this problem by conducting in vivo neuroscience experiments and developing data science tools. On the neuroscience side, they investigate functional organization and dynamic coordination in brain by combining in vivo multi-scale electrophysiology and optogenetics in rodents. This multi-modal, multi-scale approach provides the simultaneous breadth of coverage and spatio-temporal resolution required to determine neural computations at the speed-of-thought. On the data science side, they develop analysis tools for (neuro)-science, including statistical-machine learning algorithms, dynamic hierarchical models, and data standards/formats. These interpretable and predictive tools provide enhanced insight into the generative processes that produce data.