Justin Basilico

371 Followers

16 SlideShares 371 Followers 1 Following

Leading machine learning researchers and engineers to solve hard, important problems. Balancing research, software engineering, management, and vision. Doing end-to-end machine learning: creating new ideas, designing algorithms running experiments, analyzing the data, deploying in real applications, and building next-generation machine learning infrastructure to make it seamless. Currently focused on innovating the Netflix homepage through better recommendations and personalization to increase our members joy. Specialties: Machine Learning, Personalization, Recommender Systems, Deep Learning, Collaborative Filtering, Information Retrieval, Text Analysis, Artificial Intelligence

machine learning netflix personalization recommender systems recommendations large scale collaborative filtering distributed systems software engineering contextual bandits page generation causality artwork personalization online learning metrics reinforcement learning deep learning user experience experience personalization fairness ranking off policy evaluation multi-task learning multi armed bandits design patterns spark global recommendations future time failure injection reliability

Activity
About

Justin Basilico

Presentations

Recommendation at Netflix Scale

Learning a Personalized Homepage

Learning to Personalize

Personalized Page Generation for Browsing Recommendations

Lessons Learned from Building Machine Learning Software at Netflix

Recommendations for Building Machine Learning Software

Recommendations for Building Machine Learning Software

Is that a Time Machine? Some Design Patterns for Real World Machine Learning Systems

Past, Present & Future of Recommender Systems: An Industry Perspective

Déjà Vu: The Importance of Time and Causality in Recommender Systems

Making Netflix Machine Learning Algorithms Reliable

Deep Learning for Recommender Systems

Artwork Personalization at Netflix

Recent Trends in Personalization: A Netflix Perspective

Recap: Designing a more Efficient Estimator for Off-policy Evaluation in Bandits with Large Action Spaces

Recent Trends in Personalization at Netflix