CPC G06N 5/022 (2013.01) [G06N 5/01 (2023.01); G06N 5/04 (2013.01); G06N 20/00 (2019.01)] | 10 Claims |
1. An information processing device configured to generate, by processor circuitry, 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 data for learning, the information processing device comprising:
input-node specification processor circuitry, based on the input data, configured to specify input nodes corresponding to the input data, wherein each of the input nodes is located on a corresponding one of layers from beginning to end of the learning tree structure;
reliability-index acquisition processor circuitry configured to acquire a reliability index that is obtained through the learning a predetermined set of pieces of data for learning and indicates prediction accuracy;
output-node specification processor circuitry, based on the reliability index acquired by the reliability-index acquisition processor circuitry, configured to specify, from the input nodes corresponding to the input data, an output node that is a basis of the generation of a prediction output; and
prediction-output generation processor circuitry configured to generate a prediction output, based on the data for learning that is included in the state spaces that corresponds to the output node specified by the output-node specification processor circuitry, and
wherein the reliability index comprises first errors each generated at a corresponding input node among the input nodes based on a difference between an output corresponding to the input data and a prediction output based on learned data included in the state spaces that corresponds to the corresponding input node, and,
wherein the output-node specification processor circuitry is configured to specify, as the output node, a node which is among the input nodes and for which a corresponding first error among the first errors is minimal.
|