Making Music Magic: Personalizing Explainable Recommendations on Home
At Spotify, our mission is to unlock the potential of human creativity – by giving a million creative artists the opportunity to live off their art and billions of fans the opportunity to enjoy and be inspired by it. However, by delivering a seemingly unlimited selection of music, we present users with the impossible challenge of navigating this immense collection. Our approach to solving this problem is to anticipate the listening needs of our users while also showcasing the relevant depth and breadth of Spotify through personalized content and explanations. Explanations in recommender systems have the goal of both providing information about recommended content and building trust through transparency. However, different users respond to explanations differently, requiring personalized explanations on top of recommended content. This talk will examine the problem of jointly optimizing for both item selection and explanation selection for such systems, and the technical challenge this problem presents.
Catie Edwards is a Machine Learning Engineer at Spotify, where she works on personalizing content on Home. Originally from Chicago, Catie received her BSE in Computer Science from the University of Michigan’s College of Engineering. She now resides in New York City. Catie is the Co-Founder of Code Squad, a non-profit organization dedicated to bringing computer science education to underrepresented middle schools across Washington, DC.