Professor of Physics & Astronomy, Ohio University
Dr. Phillips has been applying Bayesian methods to nuclear-physics problems for more than ten years. He is particularly interested in using Bayesian methods to quantify the uncertainty in effective field theory (EFT) treatments of nuclear physics. Together with Dick Furnstahl he founded the BUQEYE collaboration, which aims to use statistical tools to answer fundamental problems in the construction and application of EFTs, with particular attention to low-energy nuclear physics. This includes Bayesian parameter estimation, model checking, model selection, and experimental design. Phillips is the Principal Investigator of the BAND collaboration.
Professor of Physics, The Ohio State University
Dr. Furnstahl is a theoretical nuclear physicist specializing in the application of effective field theory (EFT) and renormalization group methods to low-energy nuclear structure and reactions. He has been applying Bayesian methods to nuclear-physics problems for many years. Together with Daniel Phillips he founded the BUQEYE collaboration, which aims to use statistical tools to answer fundamental problems in the construction and application of EFTs, with particular attention to low-energy nuclear physics. This includes Bayesian parameter estimation, model checking, model selection, and experimental design.
Distinguished University Professor of Physics,
The Ohio State University
Dr. Heinz has been applying Bayesian methods for years to the calibration of large numerical models for the dynamical evolution of highly excited forms of nuclear matter created in ultra-relativistic heavy-ion collisions. His interests focus in particular on the quantitative extraction of information on the initial state and medium properties of quark-gluon plasma produced in such collisions. Within BAND he hopes to achieve unprecedented predictive power through full theoretical and experimental uncertainty quantification on quark-gluon plasma properties.
Professor, Department of Statistics & Probability,
Michigan State University Foundation Professor,
Michigan State University
Dr. Maiti’s research expertise is in statistical theory and methods with a focus on applications to real-life data. He is developing cutting-edge statistical theory and methods encompassing diverse research fields in biomedical engineering, genetics, nuclear physics, imaging, computational modeling, uncertainty quantifications with specific emphasis on artificial intelligence and machine learning for solving complex data science problems. He is a Fellow of the American Statistical Association (ASA) and the Institute of Mathematical Statistics (IMS).
FRIB Chief Scientist and Hannah Distinguished Professor, Michigan State University
Dr. Nazarewicz is a nuclear theorist. His research invites a strong interaction between nuclear physics, many-body problem, high-performance computing, and statistics. He has been applying statistical methods of uncertainty quantification to nuclear-physics problems for more than ten years. Most of his statistical work pertains to nuclear density functional theory. It includes various applications of parameter estimation, principal component analysis, correlation analysis, Bayesian model calibration, and model averaging and mixing - especially in the context of the science program of the Facility for Rare Isotope Beams at MSU. He is particularly interested in using Bayesian model averaging and model mixing to quantify the uncertainty of nuclear models extrapolations.
Professor of Physics and FRIB Theory Alliance Managing Director, Michigan State University
Dr. Nunes is interested in using reactions to address big questions in our field, such as where are the limits of stability, how does matter organize itself and where did heavy matter come from? While some of her work focuses on developing models for reactions with exotic unstable nuclei, another important theme relates to uncertainty quantification, model comparison and experimental design. The few-body methods used rely on effective potentials between constituents that are not well known and thus are the main source of uncertainty in current reaction models. Her group uses Bayesian statistical tools to quantify the uncertainty of the model predictions, to discriminate between models and help in identifying the optimum conditions for an experiment. More details at people.nscl.msu.edu/~nunes
Assistant Professor of Industrial Engineering and Management Sciences, Northwestern University
Dr. Plumlee studies statistical methodology with a specific focus on uncertainty quantification for expensive computational models. This includes the development and deployment of statistical learning tools in high dimensional environments. Statistical procedures of scientific interest include model emulation, parameter estimation, discrepancy detection and correction. Specific interests lie in large-scale implementations and well-justified uncertainty quantification of estimates. Within BAND, his role will be in the design, use, and implementation of cutting edge statistical tools to solve complicated nuclear physics problems n the presence of multiple models.
Associate Professor of Statistics,
The Ohio State University
Dr. Pratola’s research program is focused on two areas of statistical methodology: (1) statistical models and methodology for calibrating complex simulation models to real-world observations for parameter estimation, prediction and uncertainty quantification; and (2) statistical models and methodology for computationally scalable and flexible Bayesian non-parametric regression models for high-dimensional “big data” and parallel computation. His work is motivated by applied collaborations and has worked with researchers at the National Center for Atmospheric Research, Los Alamos National Laboratories, the Biocomplexity Institute of Virginia Tech, King Abdullah University of Science and Technology and the JADS Institute. You can follow his research at www.matthewpratola.com or on twitter @MattPratola.
Professor, Theoretical Nuclear Physics, National Superconducting Cyclotron Laboratory,
Michigan State University
Dr. Pratt is a theoretical nuclear physicist specializing in the theory, phenomenology and modeling of relativistic heavy-ion collisions. The main goal of Dr. Pratt’s research is to extract bulk properties of the quark-gluon plasma (QGP) by advancing theory and modeling, and comparing models to data from the Relativistic Heavy Ion Collider at Brookhaven National Lab and from the heavy ion program at the LHC. To that end, he has spearheaded efforts to develop strategies involving model emulators based on Bayesian statistics to constrain fundamental properties of the QGP, such as the equation of state, from experiment.
Professor of Statistics & Probability,
Michigan State University
Dr. Viens directs the Actuarial Science and Quantitative Risk Analytics program, helps to develop academic programs in data science, and assists in the university’s Statistical Training and Consulting service. His core training is in probability theory and stochastic processes. He has been applying Bayesian methods to many areas of science for more than twenty years, from climate science, to agricultural economics, and to quantitative finance. In nuclear physics, he is particularly interested in using Bayesian methods to assess the efficiency and honesty of uncertainty quantification for the limits of stability in the nuclear landscape, and to develop more efficient numerical implementations. Together with his MSU colleagues Tapabrata Maiti, and Witek Nazarewicz and his team, he is working on systematic principles for Bayesian model averaging in the presence of varied nuclear models, where improved predictive power is the goal. He is also interested in orienting his BAND collaborators towards a data-informed framework of optimal experimental design based on experiment costs, likelihood of discovery, impact of discovery, and risk preferences. Viens is the lead co-PI at MSU for the BAND collaboration.
Computational Mathematician and Deputy Director, Mathematics and Computer Science Division, Argonne National Laboratory; Senior Fellow, Northwestern University
Dr. Wild is a computational mathematician whose primary research focus is developing surrogate-model-based algorithms and software for challenging numerical optimization problems. He applies these techniques for data analysis, machine learning, and the solution of nonlinear inverse and model calibration problems. In BAND, he is especially interested in the interplay among numerical optimization, Bayesian statistics, and computationally expensive simulations of nuclei. Wild is the co-director for mathematics and computer science of the NUCLEI collaboration and leads LANS, the Laboratory for Applied Mathematics, Numerical Software, and Statistics at Argonne National Laboratory.