US 12,265,555 B2
Adaptive multi-model item selection systems and methods
Jesse Anderton, Princeton Junction, NJ (US); Maryam Aziz, Princeton Junction, NJ (US); David Bourgin, Brooklyn, NY (US); and Benjamin Austin Carterette, Wilmington, DE (US)
Assigned to Spotify AB, Stockholm (SE)
Filed by Spotify AB, Stockholm (SE)
Filed on Nov. 15, 2023, as Appl. No. 18/510,079.
Application 18/510,079 is a continuation of application No. 17/552,874, filed on Dec. 16, 2021, granted, now 11,853,328.
Prior Publication US 2024/0160643 A1, May 16, 2024
Int. Cl. G06F 16/00 (2019.01); G06F 16/28 (2019.01); G06N 3/08 (2023.01)
CPC G06F 16/285 (2019.01) [G06N 3/08 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A computer-implemented method for adaptive multi-model item selection, comprising:
receiving a request from a client device among a plurality of client devices, the request including a representation of a current state of the client device;
determining a probability distribution across a plurality of models based on the current state of the client device, wherein each of the plurality of models corresponds to a particular use case and the probability distribution provides one or more use cases the client device is in;
selecting a specific use case from the probability distribution; and
obtaining a model from the plurality of models based on the specific use case.