US 12,339,915 B2
Contextual bandit model for query processing model selection
Vinesh Reddy Gudla, South San Francisco, CA (US); David Vengerov, San Jose, CA (US); and Tejaswi Tenneti, San Carlos, CA (US)
Assigned to Maplebear Inc., San Francisco, CA (US)
Filed by Maplebear Inc., San Francisco, CA (US)
Filed on Oct. 27, 2023, as Appl. No. 18/496,724.
Prior Publication US 2025/0139176 A1, May 1, 2025
Int. Cl. G06F 16/95 (2019.01); G06F 16/9532 (2019.01); G06Q 30/0201 (2023.01); G06Q 30/0282 (2023.01); G06Q 30/0601 (2023.01)
CPC G06F 16/9532 (2019.01) [G06Q 30/0201 (2013.01); G06Q 30/0282 (2013.01); G06Q 30/0635 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A computer-implemented method comprising:
receiving, from a client device, a user query for identifying one or more items by an online system, the user query described by one or more query features;
obtaining one or more contextual features describing a context of the user query, wherein the one or more contextual features comprises:
user features describing a user associated with the client device;
retailer features describing one or more retailers hosted by the online system; and
item features describing one or more items listed on the online system; and
applying a contextual bandit model to the query features and the contextual features to select a query processing model from a plurality of query processing models wherein applying the contextual bandit model, further comprises:
outputting, for each query processing model, a predicted reward to the online system for query results identified by the query processing model; and
selecting the query processing model from the plurality of query processing models based on the predicted rewards; and
applying the selected query processing model to the user query to obtain query results; and
transmitting the query results for display on the client device.