Uber Eats has become synonymous with online food ordering. With an increasing selection of restaurants and dishes in the app, personalization is quite crucial to drive growth. One aspect of personalization is a better recommendation of restaurants and dishes so users can get the right food at the right time.

Ankit Jain and Piero Molino (Uber AI Labs) detail how to augment the ranking models with better representations of users, dishes, and restaurants. Specifically, they leverage the graph structure of Uber Eats data to learn node embeddings of various entities using state-of-the-art graph convolutional networks implemented in TensorFlow and how these methods perform better than standard matrix factorization approaches for this use case.

Prerequisite knowledge

• General knowledge of deep learning with TensorFlow

What you'll learn

• Learn how to build deep learning models on graph data using graph convolutional networks to obtain better entity representations to use for recommendation
• Discover strategies to scale a model to very big datasets