Techniques & Methods
One-Shot Learning
One-shot learning challenges models to generalize from a single labeled example per class—far fewer than typical ML requires. Approaches include metric learning (learning a similarity function), meta-learning ("learning to learn"), and leveraging rich pre-trained representations.
One-shot learning is especially valuable when labeled data is expensive or rare, such as medical imaging or industrial defect detection. Modern LLMs demonstrate strong one-shot capability through in-context learning.
Authority Links
Related Terms
Techniques & Methods
Few-Shot Learning
Model's ability to generalize from only a handful of labeled examples.
Techniques & Methods
Zero-Shot Learning
Model's ability to correctly perform tasks it was not explicitly trained for.
Techniques & Methods
Transfer Learning
Leveraging knowledge learned from one task or domain to improve performance on a related one.
Techniques & Methods
One-Shot / Few-Shot
Learning paradigms where models learn from one or very few examples to perform new tasks.