US 12,008,446 B2
Conditioned synthetic data generation in computer-based reasoning systems
Christopher James Hazard, Raleigh, NC (US); Michael Resnick, Raleigh, NC (US); Ravisutha Sakrepatna Srinivasamurthy, Raleigh, NC (US); David R. Cheeseman, Worthington, OH (US); Valeri A. Korobov, Raleigh, NC (US); Martin James Koistinen, Raleigh, NC (US); and Matthew Chase Fulp, Morrisville, NC (US)
Assigned to Howso Incorporated, Raleigh, NC (US)
Filed by Howso Incorporated, Raleigh, NC (US)
Filed on Oct. 24, 2022, as Appl. No. 17/972,164.
Application 17/972,164 is a continuation of application No. 17/006,144, filed on Aug. 28, 2020, granted, now 11,669,769.
Application 17/006,144 is a continuation in part of application No. 16/713,714, filed on Dec. 13, 2019, granted, now 11,625,625.
Application 16/713,714 is a continuation in part of application No. 16/219,476, filed on Dec. 13, 2018, abandoned.
Claims priority of provisional application 63/036,741, filed on Jun. 9, 2020.
Claims priority of provisional application 63/024,152, filed on May 13, 2020.
Claims priority of provisional application 62/814,585, filed on Mar. 6, 2019.
Prior Publication US 2023/0046874 A1, Feb. 16, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06F 15/16 (2006.01); G06F 16/22 (2019.01); G06N 5/04 (2023.01); G06N 20/00 (2019.01)
CPC G06N 20/00 (2019.01) [G06F 16/2282 (2019.01); G06N 5/04 (2013.01)] 23 Claims
OG exemplary drawing
 
1. A method comprising:
receiving a request for generation of synthetic data based on a set of training data cases;
determining one or more conditions for the synthetic data;
for each synthetic data case in the synthetic data,
for each undetermined feature in the synthetic data case,
determining one or more focal training data cases from among the set of training data cases based at least in part on the one or more conditions and any already-determined value for features in the synthetic data case;
determining a value for the undetermined feature in the synthetic data case based at least in part on the focal training data cases;
using the value for the undetermined feature in the synthetic data case;
continuing to determine undetermined features until there are no more undetermined features;
providing a computer-based reasoning model for control of a controllable system, wherein the computer-based reasoning model that was determined at least in part based on the synthetic data cases in the synthetic data;
wherein the method is performed by one or more computing devices.