US 11,895,557 B2
Systems and methods for target device prediction
Kurt Thomas Soto, Ventura, CA (US); and Charles Conor Sleith, Waltham, MA (US)
Assigned to Sonos, Inc., Santa Barbara, CA (US)
Filed by Sonos, Inc., Santa Barbara, CA (US)
Filed on Nov. 5, 2021, as Appl. No. 17/453,822.
Application 17/453,822 is a continuation of application No. 16/672,280, filed on Nov. 1, 2019, granted, now 11,172,329.
Claims priority of provisional application 62/907,367, filed on Sep. 27, 2019.
Prior Publication US 2022/0124452 A1, Apr. 21, 2022
This patent is subject to a terminal disclaimer.
Int. Cl. H04W 4/02 (2018.01); H04B 17/27 (2015.01); H04B 17/318 (2015.01); G06F 3/04847 (2022.01); G08C 17/02 (2006.01); G06F 3/0488 (2022.01); G06F 3/048 (2013.01); H04N 21/2387 (2011.01); H04N 21/472 (2011.01); G06N 20/00 (2019.01); G06F 3/0481 (2022.01); G06F 3/04883 (2022.01); G06F 3/16 (2006.01); G06N 5/04 (2023.01)
CPC H04W 4/023 (2013.01) [G06F 3/048 (2013.01); G06F 3/0481 (2013.01); G06F 3/0488 (2013.01); G06F 3/04847 (2013.01); G06F 3/04883 (2013.01); G06F 3/165 (2013.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01); G08C 17/02 (2013.01); H04B 17/27 (2015.01); H04B 17/318 (2015.01); H04N 21/2387 (2013.01); H04N 21/47217 (2013.01); G08C 2201/34 (2013.01); G08C 2201/91 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A user device comprising:
one or more processors;
a touch-screen display configured to display a graphical user interface;
at least one processor; and
at least one non-transitory computer-readable medium comprising program instructions that are executable by the at least one processor such that the user device is configured to:
after receipt of context data comprising localization data that describes a location of the user device, identify a predicted stationary playback device from a plurality of stationary playback devices in a media playback system using a prediction model that takes the context data as input and predicts a predicted stationary playback device as output;
update the graphical user interface displayed on the touch-screen display based on the predicted stationary playback device;
detect, via the touch-screen display, an input identifying a target stationary playback device from the plurality of stationary playback devices;
after detection of the input identifying the target stationary playback device, cause the target stationary playback device to perform at least one operation;
generate training data based on the predicted stationary playback device and the detected input; and
update the prediction model based on the generated training data, wherein updating the prediction model comprises:
computing a loss, wherein the loss is derived from and inversely proportional to a confidence level that the predicted stationary playback device is the target stationary playback device; and
updating weights of the prediction model based on the computed loss.