Statistical data learning under privacy constraints

摘要

A computer-implemented method is provided for statistical data learning under privacy constraints. The method includes: receiving, by a processor, a plurality of pieces of statistical information relating to a statistical object and aggregating, by the processor, the plurality of pieces of statistical information so as to provide an estimation of the statistical object. Each piece of statistical information includes an uncertainty variable, the uncertainty variable being a value determined from a function having a predetermined mean. The number of pieces of statistical information aggregated is proportional to the reliability of the estimation of the statistical object.

主权项

1. A computer-implemented method for statistical data learning under privacy constraints, the method comprising:
receiving, by a processor, a plurality of pieces of uncertain statistical information relating to a statistical object from a plurality of telematics units, wherein each piece of uncertain statistical information corresponds to an uncertainty variable having been applied to a corresponding piece of actual statistical information by a respective telematics unit from which the piece of uncertain statistical information was received, wherein the uncertainty variable corresponds to noise added by the respective telematics unit according to a common function having a predetermined mean implemented by each of the plurality of telematics units; and aggregating, by the processor, the plurality of pieces of uncertain statistical information to generate an estimated value corresponding to the statistical object.