Patent · US Active

Prototypical network algorithms for few-shot learning

US10963754B1 · kind B1 · utility

4Cited by
2References
20Claims
0Family size

Assignee

Inventors

Key dates

Filing dateSep 27, 2018
Grant dateMar 30, 2021
Priority date
Expiry dateOct 17, 2038

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N3/045
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Techniques for training an embedding using a limited training set are described. In some examples, the embedding is trained by generating a plurality of vectors from a random sample of the limited set of training data classes using a layer of the particular machine learning classification model, randomly selecting samples from the plurality of vectors into a set of samples, computing at least one distance for each sampled class from a center parameter for the class using the set of samples, generating a discrete probability distribution over the classes for a query point based on distances to a center parameter for each of the classes in the embedding space, calculating a loss value for the modified prototypical network, the calculation of the loss value being for a fixed geometry of the embedding space and including a measure of the difference between distributions, and back propagating.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.