Tensorx, Inc.

Tensordro, Inc.

Sno, Inc.

Stephen C Bedard

Mark Lane

G Edward Powell

N Edward White

John M Clerici

John M Clerci

A computer-implemented method to establish a relative importance of an input parameter pin a plurality of input parameters pin a data set input to a machine learning model, the data set represented by a j row by k column matrix I, an intersection of each row with each column defining an element, the method includes for each of the plurality of parameters pin the input data set, a computer sorts columns kof the matrix I. to produce a re-ordered matrix I; the computer determines a hyper-parameter N* of sub-matrices into which may be sorted the values in a jrow of the re-ordered matrix I; the computer generates a plurality of group sub-matrices G, each of the group sub-matrices comprising a subset of columns and the jth row; the computer inputs the re-ordered matrix Iinto a fully-trained machine learning model to produce machine learning model outputs; and the computer produces normalized mean values of the outputs.

System and methods for network sensitivity analysis

A method for evaluating a relative contribution of a first group of J data sets in a collection of N data sets, wherein N&#x3e;J, includes first applying the collection of N data sets and second applying the first group of J data sets to a model and generating one or more observations O on the collection of N data sets and the first group of J data sets, including generating a N NSA curve comprising computing, using the model, an observation Oon the collection of N data sets; and generating a N&#x2212;J NSA curve for the first group of J data sets by removing the first group of J data sets from the collection of N data sets, and generating, using the model, an observation Owith the first group of J data sets removed. The method then includes generating a measure Mof contributions of the group of J data sets based on the N NSA curve and the N&#x2212;J NSA curves.

System and methods for data evaluation through network sensitivity analysis

A data evaluation method includes a processor receiving data sets N, each of which has one or more parameters, applying the data sets N to a machine learning model and generating observations on the data sets N; and executing a network sensitivity analysis (NSA) that includes generating a N NSA curve for each of k distinct parameters in the N data sets including computing an observation Owith the data sets, and generating a N&#x2212;j NSA curve for each of the of the k distinct parameters. Generating a N&#x2212;J NSA curve includes removing the jdata set from the N data sets and computing an observation O. Executing the NSA further includes determining a contribution of a jdata set based on the k N NSA curves and the k N&#x2212;j NSA curves; and computing a relative strength Sof each of the N data sets.

Growing community of inventors

Charlotte, NC, United States of America

Stephen C Bedard