Keynote Presentation (11:35
AM - 12:30 PM on Jun 5, 2023): Speeding up Metropolis
Speaker: Dr. Jeffrey Rosenthal, University of Toronto
Jeffrey Rosenthal is a professor of Statistics at the University of Toronto, specializing in Markov chain Monte Carlo (MCMC) algorithms. He received his BSc from the University of Toronto at age 20, and his PhD in Mathematics from Harvard University at age 24. He was awarded the 2006 CRM-SSC Prize, the 2007 COPSS Presidents' Award, the 2013 SSC Gold Medal, and fellowship of the Institute of Mathematical Statistics and of the Royal Society of Canada. He has published well over one hundred research papers, and five books (including the bestseller Struck by Lightning: The Curious World of Probabilities). His web site is www.probability.ca, and on Twitter he is @ProbabilityProf.
Abstract: Markov chain Monte Carlo algorithms such as the Metropolis Algorithm converge to complicated high-dimensional target distributions to facilitate sampling. The speed of this convergence is essential for practical use. In this talk, we will present several theoretical results which can help improve the convergence speed, including diffusion limits, optimal scaling, optimal proposal shape, adaptive MCMC, adversarial Markov chains, and containment. The ideas will be illustrated using the simple graphical example available at http://www.probability.ca/met, and no particular background knowledge will be assumed.
Keynote Presentation (1:30
PM - 2:25 PM on Jun 5, 2023) - Sponsored by PROMETRIKA:
Decoding Climate Vulnerability: Harnessing the Power
of Data Science and Causal Inference
Speaker: Dr. Francesca Dominici, Harvard University
Francesca Dominici, PhD is the co-Director of the Harvard Data Science Initiative, at the Harvard University and the Clarence James Gamble Professor of Biostatistics, Population and Data Science at the Harvard T.H. Chan School of Public Health. She is an elected member of the National Academy of Medicine and of the International Society of Mathematical Statistics. She is an expert in causal inference, machine learning, Bayesian statistics. She leads an interdisciplinary group of scientists with the ultimate goal of addressing important questions in environmental health science, climate change, and biomedical science. Her productivity and contributions to the field have been remarkable. Dominici has provided the scientific community and policy makers with robust evidence on the adverse health effects of air pollution, noise pollution, and climate change. Her studies have directly and routinely impacted air quality policy. Dominici has published more than 220 peer-reviewed publications and was recognized in Thomson Reuter’s 2019 list of the most highly cited researchers–ranking in the top 1% of cited scientists in her field. Her work has been covered by the New York Times, Los Angeles Times, BBC, the Guardian, CNN, and NPR. In April 2020 she has been awarded the Karl E. Peace Award for Outstanding Statistical Contributions for the Betterment of Society by the American Statistical Association. Dominici is an advocate for the career advancement of women faculty. Her work on the Johns Hopkins University Committee on the Status of Women earned her the campus Diversity Recognition Award in 1. At the T.H. Chan School of Public Health, she has led the Committee for the Advancement of Women Faculty.
Abstract: Air pollution and climate change are two sides of the same coin. Pollutants emitted in the air can lead to changes in climatic conditions. These emissions consist of greenhouse gases. Specific components of particulate matter can either warm or cool the temperature. Short-lived climate pollutants are also dangerous air pollutants that harm people, ecosystems, and agricultural productivity.
On January 6, 2023, the Environmental Protection Agency (EPA) announced a proposal to lower the National Ambient Air Quality Standard (NAAQS) for annual PM2.5 pollution from 12 μg/m3 to between 9 and 10 μg/m3, though it continues to consider other options. Data science must inform this decision.
In this talk, I will provide an overview of data science methods, including methods for causal inference and machine learning, with the lens of policy change. This is based on a large effort of analyzing a data platform of unprecedented size and representativeness. The platform includes more than 600 million observations on the health experience of over 95% of the US population over 65 years old linked to air pollution exposure and several confounders. I will also provide an overview of studies on air pollution exposure, environmental racism, wildfires, and how they can exacerbate vulnerability to COVID-19.
Swift action on reducing short-lived climate forcers such as methane, tropospheric ozone, hydrofluorocarbons, and black carbon can significantly decrease the chances of triggering severe climate tipping points.
Banquet Talk (7:00 PM on
June 5, 2023) - Sponsored by Munich RE/HSB: Causal
Inference: What's all the fuss about?
Speaker: Dr. Jennifer Hill, New York University
Jennifer Hill develops and evaluates methods to help answer the types of causal questions that are vital to policy research and scientific development. In particular she focuses on situations in which it is difficult or impossible to perform traditional randomized experiments, or when even seemingly pristine study designs are complicated by missing data or hierarchically structured data. Most recently Hill has been pursuing two intersecting strands of research. The first focuses on Bayesian nonparametric methods that allow for flexible estimation of causal models and are less time-consuming and more precise than competing methods (e.g. propensity score approaches). The second line of work pursues strategies for exploring the impact of violations of typical causal inference assumptions such as ignorability (all confounders measured) and common support (overlap). Hill has published in a variety of leading journals including Journal of the American Statistical Association, Statistical Science, American Political Science Review, American Journal of Public Health, and Developmental Psychology. Hill earned her PhD in Statistics at Harvard University in 2000 and completed a post-doctoral fellowship in Child and Family Policy at Columbia University's School of Social Work in 2002.
Hill is also the Director of the Center for Practice and Research at the Intersection of Information, Society, and Methodology (PRIISM) and Co-Director of and the Master's of Science Program in Applied Statistics for Social Science Research (A3SR). The A3SR program has a new concentration in Data Science for Social Impact. As far as we know this is the first degree granting program in Statistics or Data Science for Social Impact or Social Good in the world.
Abstract: Most researchers in the social, behavioral, and health sciences are taught to be extremely cautious in making causal claims. However, causal inference is a necessary goal in research for addressing many of the most pressing questions around policy and practice. In the past decade, causal methodologists have increasingly been capitalizing on and touting the benefits of more complicated machine learning algorithms to estimate causal effects. These methods can take some of the guesswork out of analyses, decrease the opportunity for “p-hacking,” and may be better suited for more fine-tuned tasks such as identifying varying treatment effects and generalizing results from one population to another. However, should these more advanced methods change our fundamental views about how difficult it is to infer causality? In this talk I will discuss some potential advantages and disadvantages of using machine learning for causal inference and emphasize ways that we can all be more transparent in our inferences and honest about their limitations.
Charpignon is a PhD student at the Institute for Data,
Systems, and Society (IDSS), conducting research at the
Laboratory for Information and Decision Systems (LIDS) and the
Harvard-MIT Department of Health, Sciences, and Technology
(HST). She is co-advised by Dr. Leo Celi (Harvard-MIT) and Dr.
Maimuna Majumder (Harvard Medical School, Boston Children’s
Her research interests include causal inference, agent-based modeling, and text analysis, with applications in public health. Her focus is on the development of competing risk causal inference approaches to emulate target clinical trials for Alzheimer’s disease, using observational data from Electronic Health Records in the US and UK.
Tristan Naumann is a Principal Researcher in Microsoft Research’s Health Futures working on problems related to clinical and biomedical natural language processing (NLP). His research focuses on exploring relationships in complex, unstructured healthcare data using natural language processing and unsupervised learning techniques. He is currently serving as General Chair of NeurIPS and co-organizer of the Clinical NLP workshop at ACL. Previously, he has served as General Chair and Program Chair of the AHLI Conference on Health, Inference, and Learning (CHIL) and Machine Learning for Health (ML4H). His work has appeared in KDD, AAAI, AMIA, JMIR, MLHC, ACM HEALTH, Cell Patterns, Science Translational Medicine, and Nature Translational Psychiatry.
C. Brandon Ogbunu is an Assistant Professor in the Department of Ecology and Evolutionary Biology at Yale University, and an External Professor at the Santa Fe Institute. He is a computational biologist whose research investigates complex problems in epidemiology, biomedicine, genetics, and evolution. His work utilizes a range of methods, from experimental evolution, to biochemistry, applied mathematics, and evolutionary computation all towards a refined understanding of complex systems and disease phenomena. In addition, he runs a parallel research program at the intersection of science, society, and culture. In this capacity, he writes, gives public lectures, and creates media of various kinds.
Briana Stephenson is an Assistant Professor of Biostatistics at Harvard T.H. Chan School of Public Health. She earned her mathematics degree from Massachusetts Institute of Technology. While working as a statistician for the US Food and Drug Administration and Department of Defense, she earned an MPH degree in Biostatistics at the George Washington University. This motivated her to continue on and earn a PhD in Biostatistics from the University of North Carolina at Chapel Hill. Dr. Stephenson’s research centers on methods development and applications in nutrition, cardiovascular disease epidemiology, and population health disparities.