Information processing device, method, program, ic chip and learned model

An information processing method for generating a prediction output corresponding to input data based on a learned model that is obtained by causing a learning model having a tree structure configured by a plurality of hierarchically arranged nodes each associated with a corresponding one of hierarchically divided state spaces to learn a predetermined set of pieces of to-be-learned data, the information processing method, comprising:

• an input-node specification step of, based on the input data, specifying input nodes corresponding to the input data and each located on a corresponding one of layers from beginning to end of the learning tree structure;
• a reliability-index acquisition step of acquiring a reliability index obtained through the predetermined learning processing and indicating prediction accuracy;
• an output-node specification step of, based on the reliability index acquired by the reliability-index acquisition unit, specifying, from the input nodes corresponding to the input data, an output node that is a basis of the generation of a prediction output;
• and
• a prediction-output generation step of generating a prediction output based on the to-be-learned data that is included in the state spaces that corresponds to the output node specified by the output-node specification unit.

A computer program that causes a computer to function as an information processing device that generates a prediction output corresponding to input data based on a learned model that is obtained by causing a learning model having a tree structure configured by a plurality of hierarchically arranged nodes each associated with a corresponding one of hierarchically divided state spaces to learn a predetermined set of pieces of to-be-learned data, the program, comprising:

• an input-node specification step of, based on the input data, specifying input nodes corresponding to the input data and each located on a corresponding one of layers from beginning to end of the learning tree structure;
• a reliability-index acquisition step of acquiring a reliability index obtained through the predetermined learning processing and indicating prediction accuracy;
• an output-node specification step of, based on the reliability index acquired by the reliability-index acquisition unit, specifying, from the input nodes corresponding to the input data, an output node that is a basis of the generation of a prediction output;
• and
• a prediction-output generation step of generating a prediction output based on the to-be-learned data that is included in the state spaces that corresponds to the output node specified by the output-node specification unit.

An IC chip that generates a prediction output corresponding to input data based on a learned model that is obtained by causing a learning model having a tree structure configured by a plurality of hierarchically arranged nodes each associated with a corresponding one of hierarchically divided state spaces to learn a predetermined set of pieces of to-be-learned data, the IC chip, comprising:

• an input terminal to which the input data is input;
• an input-node specification unit that, based on the input data obtained through the input terminal, specifies input nodes corresponding to the input data and each located on a corresponding one of layers from beginning to end of the learning tree structure;
• a reliability-index acquisition unit that acquires a reliability index obtained through the predetermined learning processing and indicating prediction accuracy;
• an output-node specification unit that, based on the reliability index acquired by the reliability-index acquisition unit, specifies, from the input nodes corresponding to the input data, an output node that is a basis of the generation of a prediction output;
• a prediction-output generation unit that generates a prediction output based on the to-be-learned data that is included in the state spaces that corresponds to the output node specified by the output-node specification unit;
• and
• an output terminal from which the prediction output is output.

A learned model that is obtained by causing a learning model having a tree structure configured by a plurality of hierarchically arranged nodes each associated with a corresponding one of hierarchically divided state spaces to learn a predetermined set of pieces of to-be-learned data, the learned model being configured to:

• allow input nodes corresponding to the input data and each located on a corresponding one of layers from beginning to end of the learning tree structure to be specified based on the input data;
• allow a reliability index obtained through the predetermined learning processing and indicating prediction accuracy to be acquired;
• allow an output node that is a basis of the generation of the prediction output to be specified from the input nodes corresponding to the input data based on the reliability index acquired by the reliability-index acquisition unit;
• and
• allow the prediction output to be generated based on the to-be-learned data that is included in the state spaces that corresponds to the output node specified by the output-node specification unit.