Tuesday, January 16, 2018 - 11:00am to 12:00pm
TITLE: Precision Analytics: Optimization Through Personalization
The recent increase in data and computing resource availability has made the use of data analytics more practical in practice than ever before. In particular, the ubiquity of technologies such as smart phones, wearable devices, and smart sensors has allowed for the collection of a large amount of individual level data. In contrast to traditional data analytics, which relies on wide sampling from a population to draw conclusions, this so called deep sampled data can be used through precision analytics to create customized experiences that better serve individuals and organizations. In this talk, I discuss this precision analytics framework that builds upon the fields of reinforcement learning and data driven decision making by extending their results to applications with individual level deep sampled data. One of the main applications of this framework is to systems where a decision maker is interested in applying an intervention (or policy) to affect the behavior of an individual agent or group of agents. Two key challenges that arise when analyzing such systems are that the decision maker may have scarce resources or high risk decisions that constrain how they apply their intervention, and that the decision maker may only have partial knowledge of how an agent will react to the intervention. In this talk I discuss how these challenges can be analyzed by providing predictive models that can accurately capture individual behavior, new estimation and machine learning techniques to efficiently estimate model parameters, and effective online and batch optimization methods to calculate these interventions. I will also discuss how these approaches can be implemented in practice, particularly in the precision healthcare setting.
BIO: Yonatan Mintz is a PhD candidate at the Industrial Engineering and Operations Research Department at UC Berkeley. Over the course of his PhD research he has worked on using stochastic optimization and machine learning techniques to design personalized systems particularly in the areas of healthcare and sustainability. These algorithms were incorporated into the Cal Fitness app which received coverage in the Berkeley Engineer and was shown to have positive health outcomes through a randomized controlled trial. This work has been well recognized and earned Yonatan both the Berkeley IEOR Grassi Service Science Fellowship and a finalist position in the INFORMS Health Applications Society’s Pierskalla best paper award. In terms of methodology his research interests include topics in machine learning and nonconvex optimization. Yonatan received his bachelor's degree in Industrial and Systems Engineering with a concentration in Operations Research from Georgia Tech in 2012 and graduated with highest honors. During his PhD he has also mentored students working with the Philippines California Advanced Research Institutes, worked as researcher for UC San Francisco’s Institute for Health and Aging, and taught classes in both modeling and data analytics.