Back to results
US 2019/0244686 A1Published application
Enhanced Optimization With Composite Objectives and Novelty-Diversity Selection
AssigneeCognizant Technology Solutions U.S. Corporation
InventorsHormoz Shahrzad, Daniel Edward Fink, Risto Miikkulainen
Life of this patent
Estimating the term…
Abstract
A composite novelty method approach to deceptive problems where a secondary objective is available to diversify the search is described. In such cases, composite objectives focus the search on the most useful tradeoffs and allow escaping deceptive areas. Novelty-based selection increases exploration in the focus area, leading to better solutions, faster and more consistently and it can be combined with other fitness-based methods.
Claims17
A computer-implemented method for finding a solution to a provided problem which optimizes a plurality of objectives, comprising the steps of:
- providing a computer system having a memory storing a candidate pool database identifying a pool of candidate individuals, each identifying a respective candidate solution to the provided problem;
- a computer system testing individuals from the pool of candidate individuals against a portion of training data to develop a plurality of objective values for each of the tested individuals, each of the objective values estimating the individual's level of success with respect to a corresponding one of the objectives;
- a computer system using a predefined dominance filter to select a first subset of individuals from the candidate pool database, the dominance filter being dependent upon a plurality of composite functions of the objectives, each of the composite functions being dependent on at least one of the of objectives and at least one of the composite functions being dependent on more than one of the objectives;
- a computer system procreating new individuals from a final subset of the individuals in the candidate pool database, the final subset being dependent upon the first subset;
- inserting the new individuals into the candidate pool database and repeating the steps of testing, selecting and procreating;
- and
- operating a controlled system in dependence upon at least one of the individuals from the candidate pool database.
4 more elements
A computer-implemented method for finding one or more optimal solutions to a predetermined problem wherein the one or more optimal solutions addresses a plurality of objectives, comprising:
- testing by a first computer-implemented program each candidate solution from a predetermined pool of candidate solutions against a portion of training data to develop objective values for each of the tested candidate solution, each of the objective values estimating the candidate solution's level of success with respect to a corresponding one of the plurality of objectives;
- selecting by a second computer-implemented program a first subset of candidate solutions from the candidate pool by application of a dominance filter, wherein the dominance filter compares a plurality of composite functions for each candidate solution in the first subset against other candidate solutions in the first subset, wherein the plurality of composite functions are dependent on the plurality of objectives and at least one of the composite functions being dependent on more than one of the objectives;
- selecting by a third computer-implemented program a second subset of a predetermined number of candidate solutions from the first subset of candidate solutions, wherein each of the candidate solutions in the second subset has greater average behavioral novelty among the candidate solutions in the first subset of candidate solutions, than the average behavioral novelty of all others of the candidate solutions from the first subset;
- selecting by a fourth computer-implemented program a final subset of candidate solutions from the second subset of candidate solutions, the final subset of candidate solutions containing the one or more optimal solutions to the predetermined problem.
1 more element
A computer-implemented method for finding one or more optimal solutions to a predetermined problem wherein the one or more optimal solutions addresses a plurality of objectives, comprising:
- testing by a first computer-implemented program each candidate solution from a predetermined pool of candidate solutions against a portion of training data to develop objective values for each of the tested candidate solution, each of the objective values estimating the candidate solution's level of success with respect to a corresponding one of the plurality of objectives;
- selecting by a second computer-implemented program a first subset of candidate solutions from the candidate pool by application of a dominance filter, wherein the dominance filter compares a plurality of composite functions for each candidate solution in the first subset against other candidate solutions in the first subset, wherein the plurality of composite functions are dependent on the plurality of objectives and at least one of the composite functions being dependent on more than one of the objectives;
- selecting by a third computer-implemented program a second subset of a predetermined number of candidate solutions from the first subset of candidate solutions, wherein each of the candidate solutions in the second subset has greater average behavioral novelty among the candidate solutions in the first subset of candidate solutions, than the average behavioral novelty of all others of the candidate solutions from the first subset;
- forming by a fourth computer-implemented program a third subset of candidate solutions which is a remaining subset of candidate solutions from the first subset of candidate solutions which are not selected into the second subset of candidate solutions;
- and
- substituting candidate solutions from the third subset of candidate solutions into the second subset of individuals in a manner that improves the behavioral diversity of the individuals in the second set of individuals, to form a final set of candidate solutions, the final subset of candidate solutions containing the one or more optimal solutions to the predetermined problem.
3 more elements
Dependencies are inferred from the claim text.