Webinar Series

Data Mining Society Webinar Series

Title: Learning Policies for Allocating Scarce Housing Resources to People Experiencing Homelessness

Speaker: Jun Zhuang, University at Buffalo, The State University of New York

When: 1:30 - 2:30 PM US Eastern Time,  April 19, 2023

Where: Zoom Webinar (watch replay here!)

Abstract: Social media has been increasingly utilized to spread breaking news and risk communications during disasters. Unfortunately, due to the unmoderated nature of social media platforms such as Twitter and Facebook, rumors and misinformation could propagate widely. To address the problem, we develop a machine learning framework to predict the veracity of tweets that are spread during crisis events. We also develop two game-theoretical models “Rumor Selection for Clarification” and “Learning for Rumor Clarification”, to help decide which rumor to clarify and when to clarify, respectively. This research provides novel insights on how to efficiently monitor misinformation that is spread during disasters. We will also discuss recent research on homeland security, supply chain risk management, and wildfire management. 

Bio sketch: Dr. Jun Zhuang is Morton C. Frank Professor, Director of Graduate Studies, and Director of the Decision, Risk & Data Laboratory, Department of Industrial and Systems Engineering, the University at Buffalo. Dr. Zhuang has a Ph.D. in Industrial Engineering in 2008 from the University of Wisconsin-Madison. Dr. Zhuang's long-term research goal is to integrate operations research, big data analytics, game theory, and decision analysis to improve mitigation, preparedness, response, and recovery for natural and man-made disasters. Other areas of interest include applications to health care, sports, transportation, supply chain management, sustainability, and architecture. Dr. Zhuang has been a principal investigator of over 30 research grants funded by the U.S. National Science Foundation, by the U.S. Department of Homeland Security, by the U.S. Department of Energy, by the U.S. Air Force Office of Scientific Research, and by the National Fire Protection Association.

Title: Learning Policies for Allocating Scarce Housing Resources to People Experiencing Homelessness

Speaker: Phebe Vayanos, USC.

When: March 17, 2023 from 1:30 - 2:30 US Eastern Time

Where: Zoom Webinar (Register here!

Abstract: We study the problem of allocating scarce housing resources of different types to individuals experiencing homelessness based on their observed covariates. We leverage administrative data collected in deployment to design an online policy that maximizes mean outcomes while satisfying budget and fairness requirements. We propose a policy in which an individual receives the resource maximizing the difference between their mean treatment outcomes and the resource bid price, or roughly the opportunity cost of using a resource. Our approach has nice asymptotic guarantees and is easily interpretable. We show results on real data from the Homeless Management Information System in LA: our policies improve rates of exit from homelessness by 1.2% and policies that are fair in either allocation or outcomes by race come at very low price of fairness. In addition, to help guide the discussion among stakeholders in deciding on appropriate fairness requirements to impose when allocating scarce resources,  we propose a framework for evaluating fairness in such resource allocation systems and present a set of incompatibility results that investigate the interplay between them. Notably, we show that 1) fairness in allocation and fairness in outcomes are usually incompatible; 2) policies that prioritize based on a vulnerability score will usually result in unequal outcomes across groups; and 3) policies using group membership in addition to baseline risk and treatment effects are as fair as possible given all available information.

Bio sketch: Phebe Vayanos is a WiSE Gabilan Assistant Professor of Industrial & Systems Engineering and Computer Science at the University of Southern California. She is also an Associate Director of CAIS, the Center for Artificial Intelligence in Society at USC. Her research is focused on Operations Research and Artificial Intelligence and in particular on optimization and machine learning. Her work is motivated by problems that are important for social good, such as those arising in public housing allocation, public health, and biodiversity conservation. Prior to joining USC, she was lecturer in the Operations Research and Statistics Group at the MIT Sloan School of Management, and a postdoctoral research associate in the Operations Research Center at MIT. She holds a PhD degree in Operations Research and an MEng degree in Electrical & Electronic Engineering, both from Imperial College London. She has served as a member of the ad hoc INFORMS AI Strategy Advisory Committee and as VP of Communications for the INFORMS Section on Public Sector Operations Research. She is an elected member of the Committee on Stochastic Programming (COSP) and an Associate Editor for Operations Research Letters and Computational Management Science. She is a recipient of the NSF CAREER award and the INFORMS Diversity, Equity, and Inclusion Ambassador Program Award.

Title: From Reinforcement Learning to Sequential Decision Analytics: Toward a Universal Framework for Sequential Decision Problems

Speaker: Warren B Powell, CASTLE Lab.

When: February 24, 2023 from 11:30 - 12:30 Eastern time

Where: Zoom Webinar (Watch replay here!

Abstract: Sequential decision problems are an almost universal problem class, spanning dynamic resource allocation problems, control problems, optimal stopping/buy-sell problems, active learning problems, as well as two-agent games and multiagent problems. Application settings span engineering, the sciences, transportation, health services, medical decision making, energy, e-commerce and finance, but in this talk I will emphasize applications in transportation and logistics. These problems have been addressed in the research literature using a variety of modeling and algorithmic frameworks, including (but not limited to) dynamic programming, stochastic programming, stochastic control, simulation optimization, stochastic search, approximate dynamic programming, reinforcement learning, model predictive control, and even multiarmed bandit problems. I will present a universal modeling framework that can be used for any sequential decision problem in the presence of different sources of uncertainty. I use a “model first” strategy that optimizes over policies for making decisions. I will present four (meta)classes of policies that are the foundation of any solution approach that has ever been proposed for a sequential problem, either in the research literature or used in practice (including policies that have not been invented yet).
I will close by making the case for teaching sequential decision analytics at both the undergraduate and graduate levels, including to students in fields centered on applications as well as methodology.

About the Speaker: Warren was the founder and director of CASTLE Lab, which focused on stochastic optimization with applications to freight transportation, energy systems, health, e-commerce, finance and the laboratory sciences, supported by over $50 million in funding from government and industry. He has pioneered a new universal framework that can be used to model any sequential decision problem, including the identification of four classes of policies that spans every possible method for making decisions. This is documented in his latest book with John Wiley: Reinforcement Learning and Stochastic Optimization: A unified framework for sequential decisions. He published over 250 papers, four books, and produced over 60 graduate students and post-docs. He is the 2021 recipient of the Robert Herman Lifetime Achievement Award from the Society for Transportation Science and Logistics, the 2022 Saul Gass Expository Writing Award. He is a fellow of Informs, and the recipient of numerous other awards.