To Bid or Not To Bid: Machine Learning in Ad Tech

Abstract: In Ad Tech real-time-bidding (RTB), decisions on whether and how much to bid for an incoming request (person who just opened a web browser) must be made quickly, frequently, and intelligently. Bid too little or not at all, and you will miss the opportunity to show this person a video advertisement on behalf of one of your clients. Bid too much, and you might win, but will it be profitable? That depends on whether the person engages with the video you show her, or not. And that likelihood is determined by machine learning.

Join Justin Fortier, Principal Data Scientist, and Michael Lubavin, Lead Software Engineer, to learn how ViralGains, based here in Boston, has leveraged the power of machine learning to optimize these decisions to help place targeted video advertisements to web users on behalf of our Fortune 500 clients. We will share both the data science and the software engineering / big data perspectives of how we are solving this business problem at scale in microseconds, and how we are measuring incremental performance from machine learning. We hope you can join us!

Bio: Michael is an experienced software engineer who has worked on Wall Street, in public health, and multiple venture backed start ups. Currently Michael leads the Programmatic engineering team at ViralGains. This team is responsible for the Real Time Bidding stack that buys ad placements from exchanges in real time (milliseconds) as well as the data processing and machine learning stack that ensures the videos are shown to consumers most likely to enjoy and engage with the ad.

Open Data Science Conference