US 11,657,294 B1
Evolutionary techniques for computer-based optimization and artificial intelligence systems
Christopher James Hazard, Raleigh, NC (US)
Assigned to Diveplane Corporation, Raleigh, NC (US)
Filed by Diveplane Corporation, Raleigh, NC (US)
Filed on May 12, 2021, as Appl. No. 17/318,213.
Application 17/318,213 is a continuation of application No. 15/694,586, filed on Sep. 1, 2017, granted, now 11,010,672.
This patent is subject to a terminal disclaimer.
Int. Cl. G06N 3/12 (2023.01); G06N 3/126 (2023.01)
CPC G06N 3/126 (2013.01) 20 Claims
OG exemplary drawing
 
1. A non-transitory computer readable medium storing instructions which, when executed by one or more computing devices, cause the one or more computing devices to perform operations comprising:
receiving first candidate executable code, wherein the first candidate executable code is selected at least in part based on a first fitness score for the first candidate executable code and first ploidy of the first candidate executable code is at least two;
receiving second candidate executable code, wherein the second candidate executable code is selected at least in part based on a second fitness score for the second candidate executable code and second ploidy of the second candidate executable code is at least one;
determining a ploidy for resultant executable code;
when the determined ploidy for the resultant executable code is one, crossing over a first part of the first candidate executable code and a second part of the first candidate executable code to produce resultant executable code, wherein the resultant executable code is haploid;
when the determined ploidy of the resultant executable code is three or higher, combining the first candidate executable code and the second candidate executable code to produce the resultant executable code, wherein the resultant executable code is of the determined ploidy;
determining a third fitness score for the resultant executable code;
determining whether the resultant executable code will be used as a future candidate executable code based at least in part on the third fitness score; and
when an exit condition is met, indicating the resultant executable code as evolved executable code based at least in part on the exit condition being met.