Granted Patent B2

US 12,572,810 · App. 17/835,474 · Granted Mar 10, 2026

Inventors: Elliot Meyerson (Pasadena, CA); Risto Miikkulainen (Stanford, CA); Olivier Francon (Yssingeaux, FR); Babak Hodjat (Dublin, CA); Darren Sargent (West Linn, OR)

Assignee: Cognizant Technology Solutions US Corp.

G06N3/086

Abstract

Claims (12)

1 . A computer-implemented method for automatic discovery of intervention policies (IP) to optimize one or more objectives related to an epidemiological event, comprising:

training a predictor model, P d (C, A)=O, implemented on a processor, the predictor model being configured to receive input training data, the input historical training data sets (C, A, O) including context information (C), actions (A) performed in a given context, and outcomes (O) resulting from action performed in the given context;

establishing an initial population of candidate prescriptor models, said establishing including receiving multiple known model candidates for determining intervention policies;

distilling the multiple known model candidates into a functional architecture that is compatible with an evolutionary algorithm framework;

evolving prescriptor models, P s (C)=A, implemented on a processor, wherein the prescriptor models are evolved over multiple generations using the trained predictor model as a surrogate to evolve a subset of the candidate prescriptor models in the initial population,

the evolved prescriptor models being configured to

receive context information as input data, wherein the context information includes epidemiological event data; and

output actions that optimize the one or more objectives as outcomes corresponding to the received context information, wherein the output actions include implementation of intervention policies.

2 . The computer-implemented method of claim 1 , wherein the predictor model is a machine learning model trained with supervised methods.

3 . The computer-implemented method of claim 2 , wherein the predictor model is a neural network.

4 . The computer-implemented method of claim 1 , wherein the functional architecture is a neural network.

5 . The computer-implemented method of claim 1 , wherein the one or more objectives include minimizing a number of future epidemiological events and a level of stringency of one or more interventions.

Assignments (1)

Continuity (2)

References Cited (400)