“Distributionally Robust Learning with Applications to Health Analytics”
Speaker: Yannis Paschalidis
http://sites.bu.edu/paschalidis/ (link is external)
Wednesday, April 17, 2019 – 12:00pm
Yale Institute for Network Science, 17 Hillhouse Avenue, 3rd floor Abstract: We will present a distributionally robust optimization approach to learning predictive models, using general loss functions that can be used either in the context of classification or
regression. Motivated by medical applications, we assume that training data are contaminated with (unknown) outliers. The learning problem is formulated as the problem of minimizing the worst case expected loss over a family of distributions within a certain Wasserstein ball centered at the empirical distribution obtained from the training data. We will explore the generality of this approach, its robustness properties, its ability to explain a host of “ad-hoc” regularized learning methods, and we will establish rigorous out-of-sample performance guarantees.
Beyond predictions, we will discuss methods that can leverage the robust predictive models to make decisions and offer specific personalized prescriptions and recommendations to improve future outcomes. We will provide some examples of medical applications of our methods, including predicting hospitalizations for chronic disease patients, predicting hospital length-of-stay for surgical patients, and making treatment recommendations for diabetes and hypertension. Speaker Bio: Yannis Paschalidis is a Professor and Data Science Fellow in Electrical and Computer Engineering, Systems Engineering, and Biomedical Engineering at Boston University. He is the Director of the Center for Information and Systems Engineering (CISE). He obtained a Diploma (1991) from the National Technical University of Athens, Greece, and an M.S. (1993) and a Ph.D. (1996) from the Massachusetts Institute of Technology (MIT), all in Electrical Engineering and Computer Science. He has been at Boston University since 1996. His current research interests lie in the fields of systems and control, networks, optimization, operations research, computational biology, and medical informatics.
Prof. Paschalidis’ work has been recognized with a CAREER award (2000) from the National Science Foundation, the second prize in the 1997 George E. Nicholson paper competition by INFORMS, the best student paper award at the 9th Intl. Symposium of Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt 2011) won by one of his Ph.D. students for a joint paper, an IBM/IEEE Smarter Planet Challenge Award, and a finalist best paper award at the IEEE International Conference on Robotics and Automation (ICRA). His work on protein docking (with his collaborators) has been recognized for best performance in modeling selected protein-protein complexes against 64 other predictor groups (2009 Protein Interaction Evaluation Meeting). His recent work on health informatics won an IEEE Computer Society Crowd Sourcing Prize. He was an invited participant at the 2002 Frontiers of Engineering Symposium organized by the National Academy of Engineering, and at the 2014 National Academies Keck Futures Initiative (NAFKI) Conference. Prof. Paschalidis is a Fellow of the IEEE and the founding Editor-in-Chief of the IEEE Transactions on Control of Network Systems.
YINS Distinguished Lecturer Series
“Safeguarding Privacy in Sequential Decision-Making Problems”
Speaker: John Tsitsiklis
Clarence J. Lebel Professor of Electrical Engineering and Computer Science at MIT
Wednesday, April 10, 2019 – 12:00pm
Yale Institute for Network Science | 17 Hillhouse Avenue, 3rd floor | New Haven, CT 06511
Abstract: With the increasing ubiquity of large-scale surveillance and data analysis infrastructures, privacy has become a pressing concern in many domains. We propose a framework for studying a fundamental cost vs. privacy tradeoff in dynamic decision-making problems. More concretely, we are interested in ways that an agent can take actions that make progress towards a certain goal, while minimizing the information revealed to a powerful adversary who monitors these actions. We will examine two well-known decision problems (path planning and active learning), and in both cases establish sharp tradeoffs between obfuscation effort and level of privacy. As a byproduct, our analysis also leads to simple yet provably optimal obfuscation strategies. Based on joint work with Kuang Xu (Stanford) and Zhi Xu (MIT).
Speaker bio: John Tsitsiklis is the Clarence J. Lebel Professor of Electrical Engineering and Computer Science at MIT, and a member of the National Academy of Engineering. He obtained his PhD from MIT and joined the faculty in 1984. His research focuses on the analysis and control of stochastic systems, including applications in various domains, from computer networks to finance. He has been teaching probability for over 15 years.
Amy Justice and Walther Mothes; program leaders of the VOIC Research Program, invite you to the Virus & Other Infection-associated Cancer Program Retreat being held on Tuesday, April 30, 2019 in Hope H216.
Attached is agenda and flyer for the event. SPACE IS LIMITED.
Please contact Susan Silva (firstname.lastname@example.org) with any questions.
2019 Agenda – Final Single Page – V2.pdf
Joint Biostatistics, CS, and S&DS
BIN YU, University of California, Berkeley
Date: Monday, April 01, 2019
Time: 4:00PM to 5:15PM
Dunham Lab. see map
10 Hillhouse Avenue, Rm. 220
New Haven, CT 06511
Title: Three principles of data science: predictability,
computability, and stability (PCS)
Information and Abstract:
In this talk, I’d like to discuss the intertwining importance and connections of three principles of data science in the title and the PCS workflow that is built on the three principles. The principles will be demonstrated in the context of two collaborative projects in neuroscience and genomics for interpretable data results and testable hypothesis generation. If time allows, I will present proposed PCS inference that includes perturbation intervals and PCS hypothesis testing. The PCS inference uses prediction screening and takes into account both data and model perturbations. Finally, a PCS
documentation is proposed based on Rmarkdown, iPython, or Jupyter Notebook, with publicly available, reproducible codes and narratives to back up human choices made throughout an analysis. The PCS workflow and documentation are demonstrated in a genomics case study available on Zenodo.
3:45 p.m. Pre-talk tea Dunham Lab, Suite 222, Breakroom 228
For more details and upcoming events visit our website at
You are invited to attend the 6th annual Yale Day of Data on November 30 in Sterling Memorial Library.
The Yale Day of Data brings together researchers and data experts from across the disciplines to share experiences, challenges, and best practices related to data-intensive research. If you collect, manage, analyze, interpret, or otherwise work with data, this event is for you.
This year’s theme, “Data on Earth,” is intentionally broad, to encompass data about the Earth and the environment, data that help us understand the health and lives of Earth’s inhabitants, and data with global impact.
Keynote: William Michener (Principal Investigator of DataONE, Professor and Director of e-Science Initiatives, University Libraries, University of New Mexico), “Managing Data Throughout the Research Life Cycle to Enable New Science and Support Decision Making”
Talks by Yale faculty and researchers: Tracey Meares, Dena
Schulman-Green, Karen Seto, Alan Gerber, Jessi Cisewski, Casey King, Martin Wainstein & Sophie Janaskie
Registration is now open for the 2018 Yale Day of