METHOD FOR TESTING A MACHINE LEARNING MODEL BASED ON AN ANALYSIS OF A DESCRIPTOR'S CONTRIBUTION VALUE TO THE PREDICTION

FR3149111A1Pending Publication Date: 2024-11-29GISKARD AI
0 Cites 0 Cited by

Patent Information

Application Number
FR2023005155
Authority / Receiving Office
FR · FR
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-05-24
Publication Date
2024-11-29
Patent Text Reader

Abstract

METHOD FOR TESTING A MACHINE LEARNING MODEL FROM AN ANALYSIS OF A CONTRIBUTION VALUE OF A DESCRIPTOR TO THE PREDICTION Method for testing a machine learning model comprising: Generation (GEN1) of a first graphical representation displaying at least one value of an output data (REG1, CL1) of a first model (ML1), said value produced being assigned to said digital object (OBi), said first machine learning model (ML1) processing as input an input vector comprising a set of descriptors of a digital object (OBi) and producing as output a data from a classification or regression step; said graphical representation also displaying at least one first activatable symbol; Activation (ACT1) of said symbol and generation of a first test (T1) of the first model (ML1);Calculation of a quantification of the selected digital objects whose result of the first model (ML1) is similar to the result of the first model (ML1) executed on the first digital object (OBi). FIG.1;
Need to check novelty before this filing date? Find Prior Art