January 15, 2020

237 words 2 mins read

Achieving personalization with LSTMs

Achieving personalization with LSTMs

Personalization is a common theme in social networks and ecommerce businesses. Personalization at Uber involves an understanding of how each driver and rider is expected to behave on the platform. Ankit Jain explains how Uber employs deep learning using LSTMs and its huge database to understand and predict the behavior of each and every user on the platform.

Talk Title Achieving personalization with LSTMs
Speakers Ankit Jain (Uber)
Conference Artificial Intelligence Conference
Conf Tag Put AI to Work
Location San Francisco, California
Date September 5-7, 2018
URL Talk Page
Slides Talk Slides
Video

Personalization is a common theme in social networks and ecommerce businesses. Personalization at Uber involves an understanding of how each driver and rider is expected to behave on the platform. One way to quantify future behavior is to understand the number of trips a driver or rider will make. Ankit Jain explains how Uber employs deep learning and its huge database to understand and predict the behavior of each and every user on the platform, focusing on training LSTMs for short-term trip predictions (4–6 weeks). Uber combines past engagement data of a particular driver with incentive budgets and uses a custom loss function (i.e., zero-inflated Poisson) to come up with accurate trip predictions using LSTMs. Join in to learn how predicting rider- and driver-level behaviors helps Uber find cohorts of high-performing drivers, run personalized offers to retain users, and understand deviations from trip forecasts.

comments powered by Disqus