Put AI to work
June 26-27, 2017: Training
June 27-29, 2017: Tutorials & Conference
New York, NY

Recommending products for 1.91 billion people on Facebook

Nikita Lytkin (Facebook)
2:35pm3:15pm Thursday, June 29, 2017
Impact of AI on business and society
Location: Murray Hill E/W Level: Intermediate
Secondary topics:  Machine Learning, Media, Retail and e-commerce
Average rating: *****
(5.00, 2 ratings)

Prerequisite Knowledge

  • Basic familiarity with machine learning
  • No prior experience with recommendation systems required

What you'll learn

  • Learn how Facebook uses modern supervised machine learning methods, such as factorization machines and deep neural networks, to recommend ecommerce products to nearly two billion people.


Nikita Lytkin presents a conceptualization of digital advertising as a product recommendation problem whereby ad “creatives” are constructed in real time from highly personalized product recommendations, rather than showing the same static ad creative to different audiences, as is often done in online advertising. In order for this new approach to realize its full potential and succeed in connecting customers and businesses on Facebook, the underlying recommendation system must produce highly relevant recommendations while operating at an astonishing scale. Nikita showcases the machine learning methods and systems that make it possible.

While building this recommendation engine, Facebook tackled a number of modeling and engineering challenges arising from the unique scale of the problem of being able to recommend an extensive variety of ecommerce products to nearly two billion people. Nikita discusses the strengths and limitations of commonly used approaches, such as collaborative filtering, with respect to handling real-world cold-start scenarios, where prior customer and product interaction data may not be available, and the challenges in incorporating additional customer and product metadata that is potentially relevant to the personalization task. Nikita then shares a powerful machine learning framework that addresses these limitations and achieves deep personalization at massive scale by leveraging supervised embedding models such as factorization machines and deep neural networks in combination with vector-based indexing and online retrieval methods.

Photo of Nikita Lytkin

Nikita Lytkin


Nikita Lytkin leads machine learning teams building new monetization products at Facebook. An engineering and data science leader and advisor, previously, Nikita led teams of machine learning engineers and data scientists at LinkedIn working on making the LinkedIn News Feed highly personalized and engaging for over 400 million members and building novel data products empowering educational decision making by prospective college students. Before LinkedIn, Nikita led a team of machine learning experts in computational advertising at Quantcast. The team drove double-digit increases in performance of Quantcast’s online advertising products and company revenue by developing data-driven solutions for ad delivery, fraud detection, and campaign management. Nikita has coauthored over 20 US patent applications and continues to publish in top-tier machine learning and data mining venues. He also advises companies on building data teams and products powered by machine learning and analytics. Nikita holds a PhD in computer science from Rutgers University, where his research focused on machine learning and its applications on textual and financial data.