Implementing a learning algorithm and estimation function utilizing a randomly selected prototype function with randomly set parameters

An information processing method comprising:

• inputting a feature quantity vector and an objective variable corresponding to the feature quantity vector;
• generating a basis function by randomly selecting a prototype function among a plurality of pre-stored prototype functions and randomly setting parameters of the selected prototype function for outputting a scalar quantity by mapping the feature quantity vector;
• mapping the feature quantity vector using the basis function and calculating the scalar quantity corresponding to the feature quantity vector;
• evaluating whether or not the basis function used to calculate the scalar quantity is useful for estimating the objective variable using the objective variable along with the scalar quantity and the feature quantity vector corresponding to the scalar quantity;
• generating a next-generation basis function by executing at least one operation from the selection of the useful basis function, generation of a new basis function by mutation, and crossover using the useful basis function, and random generation of the new basis function on the basis of an evaluation result by the basis function evaluation unit, by utilizing an evolutionary technique, wherein generating further comprises randomly selecting a method of generating second-generation basis functions from among a mutation method, a crossover method, and a random generation method;
• generating an estimation function for estimating the objective variable from the scalar quantity by machine learning on the basis of the scalar quantity and the objective variable corresponding to the scalar quantity using the basis function evaluated to be useful;
• and
• outputting the estimation function,
• wherein processing of generating the estimation function by the estimation function generation unit, processing of generating the next-generation basis function by the generation alternation unit, processing of calculating the scalar quantity by the scalar quantity calculation unit, and processing of evaluating the basis function by the basis function evaluation unit are iteratively executed.

A non-transitory, computer-readable medium comprising instructions for causing a computer to execute:

• an input function of inputting a feature quantity vector and an objective variable corresponding to the feature quantity vector;
• a basis function generation function of generating a basis function by randomly selecting a prototype function among a plurality of pre-stored prototype functions and randomly setting parameters of the selected prototype function for outputting a scalar quantity by flapping the feature quantity vector;
• a scalar quantity calculation function of mapping the feature quantity vector using the basis function generated by the basis function generation function and calculating the scalar quantity corresponding to the feature quantity vector;
• a basis function evaluation function of evaluating whether or not the basis function used to calculate the scalar quantity is useful for estimating the objective variable using the objective variable input by the input function along with the scalar quantity calculated by the scalar quantity calculation function and the feature quantity vector corresponding to the scalar quantity;
• a generation alternation unit for generating a next-generation basis function by executing at least one operation from the selection of the useful basis function, generation of a new basis function by mutation and crossover using the useful basis function, and random generation of the new basis function on the basis of an evaluation result by the basis function evaluation unit, by utilizing an evolutionary technique, wherein generating further comprises randomly selecting a method of generating second-generation basis functions from among a mutation method, a crossover method, and a random generation method;
• an estimation function generation function of generating an estimation function for estimating the objective variable from the scalar quantity by machine learning on the basis of the scalar quantity calculated by the scalar quantity calculation function and the objective variable corresponding to the scalar quantity using the basis function evaluated to be useful by the basis function evaluation function;
• and
• an output function of outputting the estimation function generated by the estimation function generation function,
• wherein processing of generating the estimation function by the estimation function generation unit, processing of generating the next-generation basis function by the generation alternation unit, processing of calculating the scalar quantity by the scalar quantity calculation unit, and processing of evaluating the basis function by the basis function evaluation unit are iteratively executed.