Class Prototype
In a feature space, the mean of the feature vectors of all samples belonging to a class, serving as a compact representative of that class and used in prototype learning and incremental learning.
Timeline (last 90 days)
A class prototype is the mean of all sample feature vectors of a given class in feature space, with low storage overhead and no leakage of raw data.
Prototype quality depends on the stability of the feature extractor; if the feature space drifts as new tasks are updated, old prototypes may become invalid.
All Facts (2)
A class prototype is the mean of all sample feature vectors of a given class in feature space, with low storage overhead and no leakage of raw data.
50%UnverifiedPrototype quality depends on the stability of the feature extractor; if the feature space drifts as new tasks are updated, old prototypes may become invalid.
50%