Retrieval of outcomes of precomputed models
AssigneeDassault Systemes
InventorXavier Grehant
Life of this patent
Estimating the term…
Abstract
The invention notably relates to a computer-implemented method for retrieving model outcomes in the course of an event. The method comprises providing variables of a first set of observations of similar events, providing a multiplicity of models, indexing the multiplicity of models with the variables of the first set of observations, querying a model according to one or more variables, and returning, as a result of the query, a model.
Claims11
A computer-implemented method for retrieving model outcomes in the course of an event, comprising:
- obtaining a first set of observations of similar events, each observation being a set of mathematical variables each associated with a value, similar events being events having common mathematical variables;
- obtaining a multiplicity of models;
- training, by one or more processors, the multiplicity of models based on the mathematical variables of the first set of observations, by:
- computing subsets of the observations of the first set by applying a filtering on the mathematical variables of the first set of observations, the filtering being based on the values of the mathematical variables of the observations of the first set,
- training each model of the multiplicity of models on a respective one of the subsets of the observations of the first set, and
- indexing the multiplicity of models, including indexing each model of the multiplicity of models by the values of the mathematical variables of the subset of observations used to train the model, the indexing by the mathematical variables of the subset of observations being in the form of metadata of the model such that the model is searchable based on the values of the mathematical variables of the subset:
- collecting a second set of observations representing events that are similar to the events of the first set, the observations of the second set comprising at least the same mathematical variables as the first set;
- querying for at least one model by performing a query that includes one or more mathematical variables of the second set of observations and that allows for the retrieval of a model based on a match between values of the mathematical variables used in indexing the model and values of the mathematical variables in the query;
- and
- returning, as a result of the query, the at least one model.
A non-transitory computer readable storage medium having recorded thereon a computer program that when executed by a computer causes the computer to implement a method for retrieving model outcomes in the course of an event, the method comprising:
- obtaining a first set of observations of similar events, each observation being a set of mathematical variables each associated with a value, similar events being events having common mathematical variables;
- obtaining a multiplicity of models;
- training, by one or more processors, the multiplicity of models based on the mathematical variables of the first set of observations, by:
- computing subsets of the observations of the first set by applying a filtering on the mathematical variables of the first set of observations, the filtering being based on the values of the mathematical variables of the observations of the first set,
- training each model of the multiplicity of models on a respective one of the subsets of the observations of the first set, and
- indexing the multiplicity of models, including indexing each model of the multiplicity of models by the values of the mathematical variables of the subset of observations used to train the model, the indexing by the mathematical variables of the subset of observations being in the form of metadata of the model such that the model is searchable based on the values of the mathematical variables of the subset;
- collecting a second set of observations representing events that are similar to the events of the first set, the observations of the second set comprising at least the same mathematical variables as the first set;
- querying for at least one model by performing a query that includes one or more mathematical variables of the second set of observations and that allows for the retrieval of a model based on a match between values of the mathematical variables used in indexing the model and values of the mathematical variables in the query;
- and
- returning, as a result of the query, the at least one model.
A server comprising:
- processing circuitry coupled to a memory, the memory having recorded thereon the computer program for retrieving model outcomes in the course of an event, the processing circuitry implementing the computer program by being configured to
- obtain a first set of observations of similar events, each observation being a set of mathematical variables each associated with a value, similar events being events having common mathematical variables;
- obtain a multiplicity of models;
- train, by one or more processors, the multiplicity of models based on the mathematical variables of the first set of observations, by the processing circuitry being further configured to:
- compute subsets of the observations of the first set by applying a filtering on the mathematical variables of the first set of observations, the filtering being based on the values of the mathematical variables of the observations of the first set,
- train each model of the multiplicity of models on a respective one of the subsets of the observations of the first set, and
- index the multiplicity of models, including indexing each model of the multiplicity of models by the values of the mathematical variables of the subset of observations used to train the model, the indexing by the mathematical variables of the subset of observations being in the form of metadata of the model such that the model is searchable based on the values of the mathematical variables of the subset;
- collect a second set of observations representing events that are similar to the events of the first set, the observations of the second set comprising at least the same mathematical variables as the first set;
- query for at least one model by performing a query that includes one or more mathematical variables of the second set of observations and that allows for the retrieval of a model based on a match between values of the mathematical variables used in indexing the model and values of the mathematical variables in the query;
- and
- return, as a result of the query, the at least one model.
Dependencies are inferred from the claim text.