Machine learning device from a small number of training data examples

摘要

machine learning device (60),where an input of the machine learning system (60) is suitable to accommodate at least one pair of sizes and to process to an output size (ypred) of the machine learning system (60),where the device includes a machine-readable storage medium on which are stored commands which, when executed by a computer, have caused the computer to perform a procedure with the following steps:Provide (S11) a first probability distribution (p(f)), which describes a probability distribution of a majority of heuristics (f), whereby the heuristics (f) give estimates of a characteristic size for a sensor size;Repeated repetition of steps:Random selection (S12) of one of the heuristics (f) depending on the first probability distribution (p(f));Generate (S13) training data (D) depending on the selected heuristics (f), assigning a characteristic size to a majority of sensor sizes using the heuristics (f); andLearning (S14) of the machine learning system (60) on the basis of the training data (D), whereby the sensor sizes and associated characteristics are paired to the machine learning system (60).