Skip to content. Skip to navigation
CIM Menus
 

CIM/SOCS Seminar

Building AI-powered Experiences at Facebook Scale


Aparna Lakshmiratan, Facebook
Centre for Intelligent Machines (CIM) McGill University

October 10, 2017 at  11:00 AM
Zames Room MC437

Abstract: At Facebook, we are working on AI to help power new experiences and connect users to the content and people that they care about. In this talk we will focus on some of our recent work in enabling massive scale applications through computer vision, natural language understanding, speech and core machine learning. We will give an overview of the different applications of AI across the entire family of Facebook experiences, describe how we'’ve invested in platforms to run these algorithms at Facebook scale to enable teams with varying levels of AI expertise and finally share some of the technical and organizational lessons we have learned in the process. The explosive growth of these diverse applications of machine learning in production, the continued growth in data volume, and the complexity of large-scale learning systems bring us to some new challenges at the intersection of machine learning, and systems design.

Bio: Aparna Lakshmiratan is a technical program manager in the Applied Machine Learning group at Facebook. At Facebook, Aparna works at the intersection of cutting edge ML technology and its deployment to products at massive scale. She is currently driving their program to build new algorithms for content understanding, ranking and personalization to enable new product experiences. Prior to joining Facebook, Aparna was a principal program manager at Microsoft building and shipping several products powered by machine learning including a new Click Prediction system for Ads, enhancements to the Speller and Query Alterations engine in Bing and an interactive machine learning platform at Microsoft Research. She has a PhD from the Computer Science and Artificial Intelligence Lab at MIT.