Speaker Session Video: Finding the Good, Bad, and Ugly Predictions You’re Making Everyday
Building a machine learning model that minimizes prediction error is a core skill for predictive data scientists. Picking the right cost function is usually easy enough, but understanding how well your model is predicting all the nooks and crannies of a diverse data set can help surface trends and groupings that may have gone unnoticed. Visualizing your predictions — either segmented manually or by algorithmically defined categories — can help you identify segments of your prediction population that may need some TLC, find latent variables, or even surface potential business risks.  

Speaker Session Video: Big Data in Ad Tech
For 10+ years, the online advertising industry has been heavily dependent on machine learning algorithms to optimize auction dynamics. Buyers and sellers have relied on complex analyses of petabytes of data to set prices, floors, and other rules that govern the billions of auctions per day that fuel programmatic advertising. The increasing need to optimize results has been in conflict with the need for transparency. Buyers and sellers are questioning the "black box" nature of machine learning. Tom's talk will explore the history of algorithmic optimization in advertising and how to look for a happy medium between complexity and simplicity, as well as between optimization and transparency.

Panelist - Panel Video
Will is a Senior Data Scientist at Netflix in Los Angeles, where he builds novel demand and consumption models for the movies and TV shows Netflix streams to its subscribers globally using big data and machine-learning methods. He serves as a Data Ambassador for DataKind, bringing state-of-the-art data science practices to bear on the problems facing non-profits in the health, education, and water sectors since 2013. Previously, he worked in the online digital advertising domain in New York. Will received a Ph.D. from Harvard and a B.A. from Berkeley and has conducted research at Caltech and the University of Chicago, where he specialized in gravitational lensing studies of dark matter and dark energy in the fields of astrophysics and cosmology. 

Panelist - Panel Video
Wesley is a Senior Data Scientist at Riot Games where he develops and deploys data-powered products for League of Legends.  He has helped ship a product which provides six unique discounted offers for every player of League of Legends, and more recently he has been embedded with the Player Behavior team, where he develops machine learning algorithms to better understand and detect unsportsmanlike behavior in League of Legends.  Previously, he was a Senior Software Engineer at Google working in Google Research and as part of the app understanding team within Google Play.  He graduated from the University of Arizona with a Ph.D. in Computer Science with an emphasis on Artificial Intelligence.  His research interests include developing new algorithms to improve game experiences by leveraging machine learning techniques in game environments.

Panelist - Panel Video
Andrea is a lead data scientist at DataScience.com where she designs solutions for business needs across a range of verticals (e.g., adtech, digital media, eCommerce, gaming, IoT/sensor-based systems). Previously, she worked at Boys Town National Research Hospital to improve the efficiency of clinical tests and study the role of bilingualism in perception.  She received her Ph.D. in electrical and computer engineering from the University of Illinois Urbana-Champaign. In her research, she applied data science concepts to understand how acoustic features contribute and interact in human speech recognition. She also conducted research at MIT Lincoln Laboratories, where she used a machine learning approach to understand how biomarkers interact with major depressive disorder.