| CPC G06N 3/04 (2013.01) [G06N 3/08 (2013.01); G06N 3/082 (2013.01)] | 19 Claims |
|
1. An apparatus to modify a machine learning model, the apparatus comprising:
interface circuitry;
machine-readable instructions; and
at least one processor circuit to be programmed by the machine-readable instructions to:
identify a location of a machine learning model, the location intermediate to a first layer of the machine learning model and a second layer of the machine learning model, the location identified based on whether a first size of inputs for the first layer are different from a second size of inputs for the second layer;
insert a supervised branch into the machine learning model at the identified location;
generate a first cluster of the inserted supervised branch using a first clustering technique;
generate a second cluster of the inserted supervised branch using a second clustering technique, the second clustering technique different from the first clustering technique;
join the first cluster and the second cluster to form a joint clustering block, the joint clustering block appended to an end of the supervised branch;
perform re-clustering of the joint clustering block based on a merging degree of spatial clustering and semantic clustering; and
adjust the merging degree based on a first parameter and a second parameter, the merging degree determined using a depth of a layer of a backbone network associated with the joint clustering block.
|