Amazon Technologies, Inc.

Gurbinder Gill

Nirav Praful Desai

Michael C Moore

Kyle Andrew Farrell

Zhaoqing Ma

Canku Alp Calargun

Stanislava R Vlasseva

Ilya Levner

Venkata Krishnan Ramamoorthy

Konstantinos Boulis

Prajwal Yadapadithaya

Oladapo E Bakare

Categorizing images may include training a first neural network to cluster a plurality of images to obtain a first image embedding space, wherein a vector representation is determined for each of the plurality of images based on the training, determining a vector norm value corresponding to each of the plurality of images based on the vector representation for each of the plurality of images, and identifying a first subset of the images for which a corresponding vector norm value satisfies a predetermined vector norm quality threshold. Then, a second neural network may be trained using the first subset of images to obtain a second image embedding space, and the second image embedding space may be used to categorize additional images.

Vector norm algorithmic subsystems for improving clustering solutions

A stubbing service records requests and corresponding replies for purpose of playing back the recorded replies in response to re-receiving the corresponding requests. In a record mode, the stubbing service may receive a request from a first service intended for a second service. The stubbing service may send the request onto the second service and may, in response, receive a reply to the request from the second service. The stubbing service may then pass the reply onto the first service. When in a playback mode, the stubbing service may directly serve the stored reply to the first service in response to receiving the same request intended for the second service from the first service.

Stubbing techniques in distributed-services environments

Disclosed are various embodiments for personalized online video channels. Various embodiments of the present disclosure facilitate creation and transmission of user defined video channels for viewing television programs via the Internet. The application facilitates specification of at least one user-defined channel that transmits video content items to the client. The application maintains a data store of video content items. The application selects a subset of the video content items for inclusion in the user playlist. The application determines an order of transmission of each of the video content items in the user playlist. The application facilitates specification of at least one user-defined channel.

Customization of media play

Growing community of inventors

Seattle, WA, United States of America

Gurbinder Gill