RSVP here: bit.ly/uclaits-spring19
Join us for another installment of our Spring Lecture Series on Resilient, New Mobility
Model Checking in a Time of Simulation
Have you ever wanted a model that differed greatly from reality’s data generating process? No? Me neither. This talk will focus on how we can leverage today’s cheap computational resources to check and improve our estimated models. The concept is simple. We simulate outcomes from our model, and then we compare simulated versus observed outcomes using various test statistics. Discrepancies between the simulated and observed outcomes will suggest ways of expanding our model to improve its realism according to the chosen test statistics. Through examples and exposition, we will make connections to discrete choice, machine learning, adversarial training, and more.
Speaker: Timothy Brathwaite
Timothy Brathwaite is a data scientist at Lyft. He obtained his Ph.D. in transportation engineering from the University of California (UC) at Berkeley under the supervision of Professor Joan Walker. Motivated by efforts to predict the demand for bicycling under various policy scenarios, Timothy’s dissertation made methodological improvements to discrete choice models to account for omitted roadway variables, traveler “irrationality,” and the typically low number of cyclists in household travel surveys. He was the UCCONNECT Outstanding Graduate Student of 2016. Previously, he received his Master of City Planning and Master of Science in Civil Engineering from UC Berkeley and his Bachelor of Science in Urban Studies and Planning from the University of New Orleans. Before transitioning to Lyft, Timothy has worked with transportation consulting firms, with the City of Oakland, and with bicycle non-profit agencies.
RSVP here: bit.ly/uclaits-spring19
