US 11,907,675 B2
Generating training datasets for training neural networks
Felipe Petroski Such, San Francisco, CA (US); Aditya Rawal, Sunnyvale, CA (US); Joel Anthony Lehman, San Francisco, CA (US); Kenneth Owen Stanley, San Francisco, CA (US); and Jeffrey Michael Clune, San Francisco, CA (US)
Assigned to Uber Technologies, Inc., San Francisco, CA (US)
Filed by Uber Technologies, Inc., San Francisco, CA (US)
Filed on Jan. 17, 2020, as Appl. No. 16/746,674.
Claims priority of provisional application 62/794,477, filed on Jan. 18, 2019.
Prior Publication US 2020/0234144 A1, Jul. 23, 2020
Int. Cl. G06F 40/42 (2020.01); G06F 18/214 (2023.01); G06F 18/21 (2023.01); G06F 40/58 (2020.01); G06N 3/088 (2023.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 20/10 (2022.01); G06V 20/58 (2022.01); G06N 3/045 (2023.01); G06N 3/047 (2023.01); G06N 3/084 (2023.01)
CPC G06F 40/42 (2020.01) [G06F 18/214 (2023.01); G06F 18/217 (2023.01); G06F 40/58 (2020.01); G06N 3/088 (2013.01); G06V 10/774 (2022.01); G06V 10/776 (2022.01); G06V 20/10 (2022.01); G06V 20/58 (2022.01)] 17 Claims
OG exemplary drawing
 
1. A computer implemented method comprising:
receiving a reference training dataset including one or more reference training samples;
accessing a training dataset generator model configured to generate training datasets;
accessing a learner model configured to process the generated training datasets and the reference training dataset;
in one or more iterations:
generating, by the training dataset generator model, a training dataset according to noise as input, the training dataset including one or more training samples generated by the training dataset generator model;
training the learner model using the generated training dataset by determining a first loss based on a difference between an output value predicted by the learner model and a corresponding output value generated by the training dataset generator model, wherein parameters of the learner model are adjusted based on the first loss;
determining a second loss based on execution of the trained learner model using the reference training dataset; and
adjusting parameters of the training dataset generator model based on the second loss determined based on execution of the trained learner model using the reference training dataset;
storing one or more training datasets generated by the training dataset generator model; and
storing the training dataset generator model with adjusted parameters and the trained learner model.