We built a backtesting service to better assess financial forecast model error rates, facilitating improved forecast performance and decision making.
First introduced by Uber Engineering in January 2019, Manifold is a visual debugging tool that enables users to quickly identify performance issues in machine learning models.
In 2019, Uber AI built tools and systems that leverage ML to improve location accuracy and enhance real-time forecasting, among other applications on our platform.
We share technical challenges and lessons learned while productionizing and scaling XGBoost to train distributed gradient boosted algorithms at Uber.
To accommodate additional ML use cases, Uber evolved Michelangelo's application of the Apache Spark MLlib library for greater flexibility and extensibility.
At an April 2019 meetup on ML and AI at Uber Seattle, members of our engineering team discussed three different approaches to enhancing our ML ecosystem.
Uber AI Labs releases EvoGrad, a library for catalyzing gradient-based evolution research, and Evolvability ES, a new meta-learning algorithm enabled by this library.
Uber's Marketplace simulation platform leverages ML to rapidly prototype and test new product features and hypotheses in a risk-free environment.
We spoke to Data Science Director Fran Bell about machine learning at Uber and what she finds most challenging—and rewarding—about her work.
Uber built Manifold, a model-agnostic visualization tool for ML performance diagnosis and model debugging, to facilitate a more informed and actionable model iteration process.
We sat down with Horovod project lead, Alex Sergeev, to discuss his path to open source and what most excites him about the future of Uber's distributed deep learning framework.
Metropolis-Hastings Generative Adversarial Networks (GANs) leverage the discriminator to pick better samples from the generator after ML model training is done.
Horovod, Uber's open source distributed deep learning system, enables NVIDIA to scale model training from one to eight GPUs for their self-driving sensing and perception technologies.
Samuel Zemedkun reflects on his immigrant experience and how his part-time driving through the Uber platform funded his education and inspired his decision to join the company.
Uber built Michelangelo, our machine learning platform, in 2015. Three years later, we reflect our journey to scaling ML at Uber and lessons learned along the way.
Uber developed Michelangelo PyML to run identical copies of machine learning models locally in both real time experiments and large-scale offline prediction jobs.
One-click chat, the Uber driver app's smart reply system, leverages machine learning to make in-app messaging between driver-partners and riders more seamless.
Uber Eats engineers describe how they surface restaurant recommendations in the app using multi-objective optimization to give eaters the most satisfying experience while maintaining the health of the Uber Eats marketplace.
Uber built the next generation of COTA by leveraging deep learning models, thereby scaling the system to provide more accurate customer support ticket predictions.
12Page 1 of 2