Recommending for the World
by Justin Basilico & Yves Raimond (Netflix)
The Netflix experience is driven by a number of recommendation algorithms: personalized ranking, page generation, similarity, ratings, search, etc. On the January 6th we simultaneously launched Netflix in 130 new countries around the world, which brought the total to over 190 countries. Preparing for such a rapid expansion while ensuring each algorithm was ready to work seamlessly created new challenges for our recommendation and search teams. In this talk, we will highlight the four most interesting challenges we encountered in making our algorithms operate globally and how this improved our ability to connect members worldwide with stories they’ll love.
About the Speakers
Justin Basilico is a Research/Engineering Manager for Page Algorithms Engineering at Netflix. He leads an applied research team focused on developing the next generation of algorithms used to generate the Netflix homepage through machine learning, ranking, recommendation, and large-scale software engineering. He has also developed machine learning approaches that yielded significant improvements in the personalized ranking algorithms that drive the Netflix recommendation system. Prior to Netflix, he worked on machine learning in the Cognitive Systems group at Sandia National Laboratories. He is also the co-creator of the Cognitive Foundry, an open-source software library for building machine learning algorithms and applications. He grew up in Boston, did his undergrad at Pomona College, and started doing research in Recommender Systems in 2002 while in graduate school at Brown University.
Yves Raimond is a Research Manager at Netflix, where he leads the Search & Recommendation Algorithm Engineering team: a mixed team of researchers and engineers building the next generation of Machine Learning algorithms used to drive the Netflix experience. Before that, he was a Lead Research Engineer in BBC R&D, working on information extraction from Multimedia content. He holds a PhD from Queen Mary, University of London.
The Exploit-Explore Dilemma in Music Recommendation
by Òscar Celma (Pandora)
Were The Rolling Stones right when they said, “You can’t always get what you want; but if you try sometime you get what you need”? Recommendation systems are the crystal ball of the Internet: predicting user intentions, making sense of big data, and delivering what people are looking for before they even know they want it. As opposed to more traditional recommender systems which need only to recommend a single item or set of items, Pandora’s music recommender must provide an evolving set of sequential items, which constantly keep the experience new and exciting. In this talk I will present an overview of the recommendation algorithms at Pandora; a dynamic ensemble learning system that combines musicological data and machine learning models to provide a truly personalized experience. This approach allows us to switch from a lean back experience to a more exploration mode to discover new music tailored specifically to users individual tastes.
About the Speaker
Dr. Òscar Celma is currently Director of Research at Pandora, where he leads a team of 60 data scientists and musicologists to provide the best personalized online radio experience. Òscar published a book named “Music Recommendation and Discovery: The Long Tail, Long Fail, and Long Play in the Digital Music Space” (Springer, 2010). In 2008, Òscar obtained his Ph.D. in Computer Science and Digital Communication, in the Pompeu Fabra University (Barcelona, Spain). He holds a few patents from his work on music recommendation and discovery as well as on Vocaloid, a singing voice synthesizer bought by Yamaha in 2004.
Feature Selection For Human Recommenders
by Katherine Livins (Stitch Fix)
Recommendation systems struggle to incorporate rich features, such as those derived from natural language and images. While humans can readily process this sort of information, they cannot not scale in the same way that statistical/ML models can. As a result, hybrid-algorithms that make recommendations based on the outputs of both computers and humans are becoming increasingly popular. This talk will explore novel methods for determining what features the human side of these systems should be processing. It will outline how experimental methods (borrowed from the behavioral sciences) can be used to this end, along with how the human recommendations may be improved as a result.
About the Speaker
Katherine is a Data Scientist at Stitch Fix, working on the Human Computation team under Algorithms and Analytics. Stitch Fix uses a hybrid recommendation system, which involves both computer algorithms and human ‘stylists’. Katherine works on understanding stylist behavior through data and experimentation, and then leverages what she learns to optimize their workflow. Before Stitch Fix, Katherine completed a PhD in Cognitive and Information Science where she researched how to shape human behavior and reasoning through the manipulation of attentional cues.
Considering Supplier Relations and Monetization in Designing Recommendation Systems
by Jan Krasnodebski (Expedia)
E-commerce merchants need to optimize their recommendations beyond product attributes to include supplier considerations, long-term customer experience and monetization for long term success. Product recommendations for optimizing customer conversion can be modeled effectively with predictive analytic methodologies. However, supplier and customer experience elements are not easily modeled in the same manner. This paper outlines an algorithmic approach for these considerations from Expedia’s experiences.
About the Speaker
Jan Krasnodebski, as Expedia’s Director of Lodging Revenue Optimization, built a team of machine learning experts that personalizes and optimizes hotel recommendations across Expedia’s websites. He previously was the Director of Pricing and Portfolio Management for a major North American bank and a pricing consultant.
A Cross-Industry Machine Learning Framework with Explicit Representations
by Nathan R. Wilson (Nara Logics)
At Nara Logics, we provide recommendations for ecommerce, supply chain, financial services, travel & hospitality, operations and more for the Global 200. We’ve learned that for machine intelligence to be accepted, it must interact seamlessly with humans, expose its reasoning to humans, and even incorporate human feedback in real time into its decision making. Just as you take your friends’ recommendations more seriously when you can probe their mental model of your likes and dislikes, machine recommendations are more appealing when users understand how they were generated and can provide feedback to those recommendations. These aspects are necessary as commercial interfaces increasingly leverage recommendations alongside statistical analysis.
In this talk, I will discuss a new general purpose framework, “synaptic intelligence”, for learning and maintaining explicit representations in real-time. This framework can be used both to make recommendations and to explain why each recommendation was made to meet the objectives outlined. We will also discuss how we extend our technology to enable users to have meaningful interactions where they’re able to find what they’re looking for. We believe in developing recommendation technology that facilitates a conversation with the user. The examples we showcase will demonstrate how we are able to leverage an explicit representation to make this possible.
About the Speaker
Nathan Wilson is a scientist and entrepreneur focused on actualizing powerful new models of brain-based computation. After many years at MIT working on the mathematical logic of neural circuits, Nathan co-founded Nara Logics, a Cambridge, MA artificial intelligence company developing “synaptic intelligence” that automatically finds and refines connections across data for recommendations and decisions within enterprises. Nathan holds many patents in AI and his research has been featured in top journals including Nature, Science, Proceedings of the National Academy of Sciences, Neuron, and the MIT Press. An enthusiastic writer and teacher who routinely appears in the popular press on current topics in AI, Nathan now works to guide advancements at Nara Logics as CTO.
