CPC G06N 20/00 (2019.01) | 20 Claims |
10. A system for classifying objects by a machine learning (ML) model, the system comprising:
one or more processors; and
a memory storing instructions that, when executed by the one or more processors, causes the system to perform operations comprising:
obtaining an object to be classified by the ML model;
providing the object to a plurality of layer level classification models of a layer of the ML model;
for each of the plurality of layer level classification models of the layer:
generating a classification for the object and an uncertainty metric corresponding to the classification by the layer level classification model; and
providing the classification and the uncertainty metric to a meta-model of the ML model; and
generating, by the meta-model, a final classification of the object and a confidence in the final classification based on the plurality of classifications and the plurality of uncertainty metrics corresponding to the plurality of classifications.
|