US 12,456,075 B2
Virtual imputation learning for one-sided material properties
Jens Strabo Hummelshøj, Brisbane, CA (US)
Assigned to Toyota Research Institute, Inc., Los Altos, CA (US)
Filed by Toyota Research Institute, Inc., Los Altos, CA (US)
Filed on Jan. 24, 2022, as Appl. No. 17/582,638.
Prior Publication US 2023/0237365 A1, Jul. 27, 2023
Int. Cl. G06N 20/00 (2019.01); G06F 18/211 (2023.01); G16C 60/00 (2019.01)
CPC G06N 20/00 (2019.01) [G06F 18/211 (2023.01); G16C 60/00 (2019.02)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a processor; and
a memory communicably coupled to the processor and storing an acquisition module and machine learning (ML) module with machine-readable instructions that, when executed by the processor, cause the processor to:
select a training data set from a candidate data set via the machine-readable instructions of the acquisition module, the training data set including a first subset of material candidates each having a material property within a predefined range and a second subset of material candidates each having a one-sided material property outside the predefined range; and
via the machine readable instructions of the ML module:
impute a fixed value for the material property outside the predefined range for the second subset of material candidates;
train a ML model with the first subset of material candidates and the second subset of material candidates with the imputed fixed value; and
predict, using the trained ML model, the material property for materials having the material property within the predefined range and materials having the one-side material property outside the predefined range.