ORC IAP Seminar 2016
"Analytics in Operations Research"
Description: There is little doubt that "Analytics" is becoming a major part of Operations Research, gaining a lot of interest in the research community as well as in a variety of industries. This year, the MIT Operations Research Center will welcome speakers from various fields to talk about the role and impact of Analytics in improving decision making. Four speakers, from both industry and academic settings, will present their work and experience on topics ranging from marketing to sharing economy.
9:30am - 10:00am - Intro and Continental Breakfast
Speaker: John Silberholz - Lecturer in the MIT Sloan School of Management
Title: An Analytics Approach to Designing Combination Drug Therapies for Cancer
Abstract: We present a data-driven approach for designing new drug therapies for advanced gastric and breast cancers. Our approach combines (i) construction of a large-scale database of clinical study results, (ii) statistical modeling to predict outcomes of new drug combinations, and (iii) optimization models to select novel treatments that strike a balance between maximizing patient outcomes and learning new information about treatments that may be useful in designing future therapies.
Bio: John Silberholz is a lecturer and postdoctoral fellow in the MIT Operations Research and Statistics group. He is broadly interested in data-driven decision making, especially in the healthcare space. His current research focus is on designing effective population screening strategies for detecting cancer and on using information from published clinical trials to better treat cancer patients and to better design further clinical trials. John received his PhD from the MIT Operations Research Center in 2015 and bachelor's degrees in math and computer science from the University of Maryland in 2010.
Speaker: Dean Eckles - Assistant Professor in the MIT Sloan School of Management - Marketing Group
Title: Estimating Effects in Networks with Peer Encouragement Designs
Abstract: Interactions among humans enable the spread of information, preferences, behavior, and disease. Despite large-scale measurement of human behaviors, credible identification and estimation of peer influence effects remains difficult. Even well-designed observational studies that adjust for thousands of covariates can have remaining bias from the confounding of influence and homophily. I review attempts to use randomized experiments and observational studies to estimate peer effects in networks, including issues of both causal and statistical inference. I then focus on peer encouragement designs, a class of experimental designs for identifying peer effects that randomly assign a focal individual’s peers to encouragements to specific behaviors that affect the individual. We illustrate this method by reporting on a large peer encouragement design on Facebook that allows estimating the effects of receiving feedback from peers on posts shared by focal individuals. This is joint work with René Kizilcec and Eytan Bakshy.
Bio: Dean Eckles is a social scientist, statistician, and assistant professor in the MIT Sloan School of Management. He was previously a member of the Core Data Science team at Facebook. He studies how interactive technologies affect human behavior by mediating, amplifying, and directing social influence — and the statistical methods to study these processes. Dean’s current work uses large field experiments and observational studies. His research appears in peer reviewed proceedings and journals in computer science, marketing, and statistics. Dean holds degrees from Stanford University in philosophy (BA), cognitive science (BS, MS), and statistics (MS), and communication (PhD).
Speaker: Jae-wook Ahn - Data Scientist at IBM Watson Life, TJ Watson Research Center
Title: Understanding Chef Watson Users
Abstract: IBM Chef Watson is a cognitive computing platform developed by IBM Watson Life to let users dynamically generate recipes based on their choice of ingredients, dish, and styles. We collect Chef Watson diverse user information from various sources and try to understand their preferences, interests, and behavioral characteristics in order to provide smarter applications. In this seminar I show the strategy to collect, maintain, and analyze the user data. I also present a use-case to visually analyze user activities and the lessons we learned from it.
Bio: Jae-wook Ahn is a Watson Data Scientist at IBM Watson Life. He received his Ph.D. in information science from University of Pittsburgh and worked as a postdoctoral NSF-CRA Computing Innovation Fellow at the Human-Computer Interaction Lab (HCIL), University of Maryland at College Park. His research interests include user modeling, personalization, recommender systems, and visual analytics for intelligent systems.
Speaker: Jon Petersen - Senior Data Scientist at Uber
Title: Marketplace Optimization at Uber
Abstract: The rapid acceleration of the sharing economy has introduced a myriad of challenges for two-sided marketplaces. This talk will address how optimization and machine learning are powering the dynamic marketplace at Uber, a platform that has connected over one billion riders and drivers across more than 60 countries. Topics that will be surveyed include dynamic pricing, matching riders in uberPOOL, and real-time on-demand delivery services.
Bio: Jon Petersen is a Senior Data Scientist at Uber Technologies in San Francisco, CA. Dr. Petersen’s work has applied large-scale and mixed-integer optimization applications to near real-time transportations systems, particularly within ridesharing and aviation. Prior to joining Uber he has worked on developing algorithms used by airlines to recovery from irregular operations. He received his PhD in Industrial and Systems Engineering from Georgia Tech in 2012.
If you have any questions, please contact Amine Anoun <firstname.lastname@example.org>, Arthur Flajolet <email@example.com>, or Zachary Clayton Saunders <firstname.lastname@example.org> by email.
Independent Activities Period (IAP)
The OR Center participates in MIT's Independent Activities Period (IAP) by offering a series of informational seminars focusing on the OR Center and on current research and the practice of OR. IAP is a month-long period between the fall and spring terms (usually the month of January) during which all members of the MIT community participate in developing individual interests for the benefit of the community and themselves.
A listing of all IAP activities sponsored by the ORC can be found here.