US 12,307,375 B2
Systems and methods for parameter optimization
Sebastien Ouellet, Ottawa (CA); Phillip Williams, Ottawa (CA); Nathaniel Stanley, Ottawa (CA); Jeffery Downing, Ottawa (CA); and Liam Hebert, Ottawa (CA)
Assigned to KINAXIS INC., Ottawa (CA)
Filed by Kinaxis Inc., Ottawa (CA)
Filed on Jan. 5, 2024, as Appl. No. 18/405,116.
Application 18/405,116 is a continuation of application No. 17/973,201, filed on Oct. 25, 2022, granted, now 11,900,259.
Application 17/973,201 is a continuation of application No. 16/865,707, filed on May 4, 2020, granted, now 11,514,328, issued on Nov. 29, 2022.
Prior Publication US 2024/0144018 A1, May 2, 2024
Int. Cl. G06N 3/086 (2023.01); G06F 16/901 (2019.01); G06N 3/126 (2023.01)
CPC G06N 3/086 (2013.01) [G06F 16/9027 (2019.01); G06N 3/126 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for data configuration of a data set for use in a machine learning system, the method comprising:
defining (i) one or more objectives and (ii) one or more parameters for evaluating each of the one or more objectives;
generating an initial set of parameters to produce an initial configuration of the data set;
evaluating a fitness function of each of the one or more objectives based on the initial set of parameters;
obtaining an initial Pareto Front comprising a plurality of objective points, each Pareto objective point associated with the fitness function;
applying recursively, a genetic algorithm to the plurality of Pareto objective points, thereby generating new sets of objective points forming one or more new Pareto Fronts, until the initial Pareto Front converges with a final Pareto Front forming a converged Pareto Front;
generating recommended configurations for the data set based on objective points on the converged Pareto Front, wherein the objective points on the converged pareto Front represent the one or more objectives associated with the one or more parameters;
selecting one or more of the recommended configurations;
configuring the data set using the one or more recommended configurations for use in the machine learning system;
and operating the machine learning system based on the configured data set.