Patent · US Active

Regularized neural network architecture search

US11144831B2 · kind B2 · utility

3Cited by
0References
16Claims
0Family size

Assignee

Inventors

Key dates

Filing dateJun 19, 2020
Grant dateOct 12, 2021
Priority date
Expiry dateJun 19, 2040

Classification

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

Abstract

A method for receiving training data for training a neural network (NN) to perform a machine learning (ML) task and for determining, using the training data, an optimized NN architecture for performing the ML task is described. Determining the optimized NN architecture includes: maintaining population data comprising, for each candidate architecture in a population of candidate architectures, (i) data defining the candidate architecture, and (ii) data specifying how recently a neural network having the candidate architecture has been trained while determining the optimized neural network architecture; and repeatedly performing multiple operations using each of a plurality of worker computing units to generate a new candidate architecture based on a selected candidate architecture having the best measure of fitness, adding the new candidate architecture to the population, and removing from the population the candidate architecture that was trained least recently.

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