| CPC G06N 20/00 (2019.01) [G06F 3/013 (2013.01); G06T 11/00 (2013.01)] | 20 Claims |
|
1. A method, performed by processing circuitry of one or more computing devices, of operating a mixed reality graphical environment (MRGE), the method comprising:
for each of a plurality of training instances of the MRGE running on the processing circuitry, monitoring usage data of a first plurality of users of that training instance of the MRGE;
for each of the plurality of training instances, calculating a first set of parameters that describe usage of that training instance of the MRGE using the monitored usage data for that training instance;
training a machine learning (ML) model on the calculated set of parameters for each training instance of the MRGE with an associated supervisory signal that indicates one of a set of usage scenarios for each training instance of the plurality of training instances;
monitoring usage data of a second plurality of users of a non-training instance of the MRGE;
calculating a second set of parameters that describe usage of the non-training instance of the MRGE using the monitored usage data for the non-training instance of the MRGE;
feeding the calculated second set of parameters that describe usage of the non-training instance of the MRGE into the trained ML model to obtain a usage scenario of the non-training instance of the MRGE; and
adjusting features of the non-training instance of the MRGE based on the usage scenario.
|