US 11,941,527 B2
Population based training of neural networks
Maxwell Elliot Jaderberg, London (GB); Wojciech Czarnecki, London (GB); Timothy Frederick Goldie Green, London (GB); and Valentin Clement Dalibard, London (GB)
Assigned to DeepMind Technologies Limited, London (GB)
Filed by DeepMind Technologies Limited, London (GB)
Filed on Mar. 13, 2023, as Appl. No. 18/120,715.
Application 18/120,715 is a continuation of application No. 16/766,631, granted, now 11,604,985, previously published as PCT/EP2018/082162, filed on Nov. 22, 2018.
Claims priority of provisional application 62/590,177, filed on Nov. 22, 2017.
Prior Publication US 2023/0281445 A1, Sep. 7, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/08 (2023.01); G06N 3/048 (2023.01)
CPC G06N 3/08 (2013.01) [G06N 3/048 (2023.01)] 20 Claims
OG exemplary drawing
 
1. A method of training a neural network having a plurality of network parameters to perform a particular neural network task and to determine trained values of the network parameters using an iterative training process having a plurality of hyperparameters, the method comprising:
maintaining a plurality of candidate neural networks and, for each of the plurality of candidate neural networks, data specifying: (i) respective values of the network parameters of the candidate neural network, (ii) respective values of the hyperparameters of the candidate neural network, and (iii) a quality measure that measures a performance of the candidate neural network on the particular neural network task;
for each of the plurality of candidate neural networks, repeatedly performing the following training operations, comprising:
repeatedly updating the values of the network parameters of the candidate neural network in accordance with the maintained values of the hyperparameters of the candidate neural network, and updating the quality measure of the candidate neural network based on the updated values of the network parameters of the candidate neural network until termination criteria are satisfied;
in response to determining that the termination criteria are satisfied, setting (i) the values of the hyperparameters of the candidate neural network and (ii) the updated values of the network parameters of the candidate neural network, the setting comprising:
sampling another candidate neural network model from the plurality of candidate neural networks, wherein the other candidate neural network is associated with data specifying maintained quality measure of the other candidate neural network, and
setting (i) the values of the hyperparameters of the candidate neural network and (ii) the updated values of the network parameters of the candidate neural network based on the updated quality measure of the candidate neural network and the maintained quality measure of the other candidate neural network to obtain new values;
updating the maintained data of the candidate neural network to specify the new values of the hyperparameters, the new values of the network parameters, and the new value of the quality measure after the setting; and
selecting the trained values of the network parameters from parameter values in the maintained data based on the maintained quality measures of the plurality of candidate neural networks after the training operations have repeatedly been performed.