Glosarium AI

📖

istilah

Few-Shot Learning (FSL)

Machine learning paradigm where a model learns to recognize new classes from a very small number of training examples, typically between 1 and 5 samples per class.

📖

istilah

N-way K-shot Learning

Formalization of the few-shot problem where N represents the number of classes to discriminate and K the number of available examples per class for learning.

📖

istilah

Feature Reweighting

Few-shot technique that dynamically adjusts the importance of extracted features based on support examples to better adapt to new classes.