Lyft, Inc.

Wei You

Limin Shen

Michael Francis Lodick

Jeff Pinner

Joan Devassy

Nadha Nafeeza Gafoor

Ritwik Subir Das

Swarn Avinash Kumar

John Michael Sparks

Kia Nishimine

Angela Lam

Wiebke Poerschke

Hristo Stefanov Stefanov

Aahel Iyer

Embodiments provide approaches to selectively integrate with various providers based on reliability factors, user preferences, and/or other criteria. For example, a user in a location where an on-demand provider is available may attempt to utilize services provided by the on-demand provider in a different location where the on-demand provider has limited or no services available, but where one or more other on-demand providers may be available. The user may attempt to access services provided by the on-demand provider using an application provided by the on-demand provider. However, in the situation where the user is in the location where services provided by the on-demand provider are not available, the user would have to secure other means of accessing those services. In such a situation the user may be able to utilize the application to request services with at least one partner on-demand provider. This can include, for example, enabling the user to use the application to access such services, and thus maintain a familiar front-end user experience, while on the backend a flexible application program interface (API) framework is utilized to enable requests to partner on-demand providers who can satisfy requests on behalf of the on-demand provider.

Flexible API framework

This disclosure describes a card-scan system that can update a card-scan machine learning model to improve card-character predictions for payment cards, driver licenses, or other character-bearing cards by using an active-learning technique that learns from card-scan representations indicating corrections by users to predicted card characters. In particular, the disclosed systems can use a client device to capture and analyze a set of card images of a character-bearing card to predict card characters using a card-scan machine learning model. The disclosed systems can further receive card-scan gradients representing one or more corrections to incorrectly predicted card characters. Based on the card-scan gradients, the disclosed systems can generate active-learning metrics and retrain or update the card-scan machine learning model based on such active-learning metrics. By utilizing and updating the card-scan machine learning model, the disclosed systems can improve the accuracy with which card-character-detection systems predict card characters while preserving data security and verifying the presence of a physical character-bearing card.

Automatically updating a card-scan machine learning model based on predicting card characters

Growing community of inventors

San Francisco, CA, United States of America

Wei You