Disclosed herein is an improved mechanism for dynamic format selection of media content. The mechanism can use a machine learning model to select a format (e.g., a playback resolution) for providing media content based on information available at the time when the request for providing the media content is received. Such information can include, for example, location information (e.g., the geographic location of a user device that is presenting the media content), account information (e.g., previously selected media content items, watch time statistical information, subscription information, etc.), and/or network condition information (e.g., current egress traffic from one or more servers associated with the media streaming service).

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.