INFORMATION PROCESSING APPARATUS

PROBLEM TO BE SOLVED: To provide a high-value-added information processing apparatus configured to execute various simulations easily at a low cost.SOLUTION: An information processing apparatus 1 includes: state data acquisition means 3 for acquiring state data and state transition speed data from a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: KIN TENKAI, MATSUBAYASHI YURIKO, OGASAWARA TOSHIKI, NUMAKURA AKIO, MIYOSHI TAMOTSU, AKASHI TAKUYA, TOMIZAWA TAKEYA
Format: Patent
Sprache:eng ; jpn
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:PROBLEM TO BE SOLVED: To provide a high-value-added information processing apparatus configured to execute various simulations easily at a low cost.SOLUTION: An information processing apparatus 1 includes: state data acquisition means 3 for acquiring state data and state transition speed data from a system 2; tree structure management means 5; and node data management means 7. The tree structure management means 5 includes tree structure data 6. The tree structure data has a hierarchical lattice structure. The node data management means 7 includes node data 8. Data on the state and/or state transition speed of the system 2 input in the state data acquisition means 3 are input to the tree structure management means 5 and the node data management means 7. A relationship of the state and/or state transition speed are approximated by a function, on the basis of the tree structure data 6 and the node data 8, to learn the state and/or state transition speed. The state and/or state transition speed are estimated by learning, to output predicted values 9 of the state and/or state transition speed.SELECTED DRAWING: Figure 1 【課題】種々のシミュレーションが容易にかつ低コストで実行可能な付加価値の高い情報処理装置を提供する。【解決手段】情報処理装置1は、対象システム2から状態データ及び状態遷移速度データを取得する状態データ取得手段3と、木構造管理手段5と、ノードデータ管理手段7とを含み、木構造管理手段5は木構造データ6を備え、該木構造データは階層型の格子構造を有しており、ノードデータ管理手段7はノードデータ8を備え、状態データ取得手段3に入力された対象システム2の状態及び/又は状態遷移速度に関するデータを木構造管理手段5及びノードデータ管理手段7に入力し、状態及び/又は状態遷移速度の関係性を木構造データ6及びノードデータ8に基づいて関数により近似することで状態及び/又は状態遷移速度を学習し、学習により状態及び/又は状態遷移速度を予測し、状態及び/又は状態遷移速度の予測値9を出力する。【選択図】図1