A/B Testing and Beyond: Experimental Design and Popular ML Algorithms in Online Attribution In this session Katie Sasso from the Columbus Collaboratory will discuss applications of experimental design methods and popular machine learning algorithms, including eXtreme Gradient Boosting, as applied to online attribution. The talk will include a high level overview of: the statistical underpinnings of such algorithms, pros and cons of applying them in the context of online attribution, and different experimental design strategies for testing campaign effects that extend beyond the traditional A/B paradigm. https://drfgumv0s4lyk.cloudfront.net/wp-content/uploads/2018/01/sasso_headshot_linkedin_redsize-267×300.jpgKatie Sasso is a Data Scientist at the Columbus Collaboratory. Katie received her Ph.D. in Experimental Psychology from The Ohio State University in 2017. During her Ph.D. program Katie worked with the Nationwide Center for Advanced Customer Insights (NCACI) and used her research and statistics skills to provide data-driven insights and solutions to strategic questions within Nationwide Insurance. Since that time she has been applying her training in statistics and experimental design to a broad range of problems across the many different industries represented by the Columbus Collaboratory’s member companies. Her project work spans a variety of use cases from predictive modeling and natural language processing to automation and UI building. All of our Columbus WAWs are free of charge thanks to the generosity of our sponsors: A /V support, including a Facebook Livestream of the event, provided by Darin Young from .