NOTE: DUE TO A SCHEDULING CONFLICT WE WILL BE MEETING TONIGHT IN ROOM L324 EECS Conference Room IN TECH - MCCORMICK SCHOOL OF ENGINEERING. A MAP TO THE BUILDING IS GIVEN ABOVE. PLEASE CONTACT JEREMY AT 317.417.5176 ANYTIME IF YOU NEED HELP LOCATING THE BUILDING.
We're incredibly excited to host Halla Yang as this month's speaker at NU ML! Come out and learn what it takes to succeed in a data mining competition hosted by Kaggle.com (http://www.kaggle.com) from a Kaggle-champion!
Title: How to Compete on Kaggle and Win
Kaggle (http://www.kaggle.com/) is an innovative solution for statistical/analytics outsourcing. Companies, governments and researchers present datasets and problems - the world's best data scientists then compete to produce the best solutions. At the end of a competition, the competition host pays prize money in exchange for the intellectual property behind the winning model.
I will provide an overview of Kaggle and talk about some of the tools and techniques I have used to win my first Kaggle competition. See http://www.kaggle.com/users/36141/halla#profile-results for a complete set of my competition results.
About the Speaker I hold a B.A. in physics from Harvard, as well as a Ph.D. in Business Economics from Harvard. After finishing my doctorate, I developed trading strategies at Goldman Sachs Asset Management in New York and managed portfolios at Arrowstreet Capital in Boston. When my wife accepted a tenure-track position at Northwestern, I resigned from my job to move to Evanston. My former employer required me to sit out for one year, and I decided to spend that year exploring a topic I knew little about but was keenly interested in - machine learning - via competing on Kaggle. Eight competitions later, I am currently ranked in the top 50 out of over 76,000 registered Kaggle users, and recently won my first competition, an invitation-only event sponsored by Pfizer. I have a Kaggle hoodie to boot. This March, when my non-compete is over, I'll be joining a proprietary trading firm in Chicago.