Center for Applied Mathematics Colloquium
Abstract: In this talk I will go over the core concepts and intuition behind Bayesian Optimization as a black-box, global optimization technique and tradeoffs between other approaches. I will explain how techniques like these are applied in a variety of industries and applications from algorithmic trading firms to streaming services and beyond. We'll explore the research and intuition behind some of the applied research that needs to be done when extending these traditional methods to attack a variety of real world problems. I'll talk about my journey over the last decade from a CAM student working on these problems to writing open source code to productionize them in industry to forming SigOpt, a software company that produces model optimization software used around the world in industry, academia, and government. I will talk about the journey from academic open source to an enterprise software company and the tradeoffs of building a startup from active research and collaborating with some of the top modeling firms in the world. Bio: Scott has been applying optimal learning techniques in industry and academia for years, from bioinformatics to production advertising systems. He is co-founder and CEO of SigOpt, a 30 person San Francisco based startup that provides a model experimentation and optimization platform to customers worldwide. Before SigOpt, Scott worked on the Ad Targeting team at Yelp leading the charge on academic research and outreach with projects like the Yelp Dataset Challenge and open sourcing MOE. Scott holds a PhD in Applied Mathematics and an MS in Computer Science from Cornell University and BS degrees in Mathematics, Physics, and Computational Physics from Oregon State University. Scott was chosen as one of Forbes’ 30 under 30 in 2016.