Patent · US Active

Annealed dropout training of neural networks

US10373054B2 · kind B2 · utility

2Cited by
1References
19Claims
0Family size

Assignee

Inventors

Key dates

Filing dateSep 1, 2015
Grant dateAug 6, 2019
Priority date
Expiry dateJul 20, 2037

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N20/00
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

Systems and methods for training a neural network to optimize network performance, including sampling an applied dropout rate for one or more nodes of the network to evaluate a current generalization performance of one or more training models. An optimized annealing schedule may be generated based on the sampling, wherein the optimized annealing schedule includes an altered dropout rate configured to improve a generalization performance of the network. A number of nodes of the network may be adjusted in accordance with a dropout rate specified in the optimized annealing schedule. The steps may then be iterated until the generalization performance of the network is maximized.

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