Processing

Please wait...

Settings

Settings

Goto Application

1. WO2019186996 - MODEL ESTIMATION SYSTEM, MODEL ESTIMATION METHOD, AND MODEL ESTIMATION PROGRAM

Publication Number WO/2019/186996
Publication Date 03.10.2019
International Application No. PCT/JP2018/013589
International Filing Date 30.03.2018
IPC
G06Q 10/04 2012.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
10Administration; Management
04Forecasting or optimisation, e.g. linear programming, "travelling salesman problem" or "cutting stock problem"
G06Q 30/02 2012.01
GPHYSICS
06COMPUTING; CALCULATING OR COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
30Commerce, e.g. shopping or e-commerce
02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
CPC
G06Q 10/04
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
10Administration; Management
04Forecasting or optimisation, e.g. linear programming, "travelling salesman problem" or "cutting stock problem"
G06Q 30/02
GPHYSICS
06COMPUTING; CALCULATING; COUNTING
QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
30Commerce, e.g. shopping or e-commerce
02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
Applicants
  • 日本電気株式会社 NEC CORPORATION [JP]/[JP]
Inventors
  • 江藤 力 ETO Riki
Agents
  • 岩壁 冬樹 IWAKABE Fuyuki
  • 塩川 誠人 SHIOKAWA Masato
Priority Data
Publication Language Japanese (JA)
Filing Language Japanese (JA)
Designated States
Title
(EN) MODEL ESTIMATION SYSTEM, MODEL ESTIMATION METHOD, AND MODEL ESTIMATION PROGRAM
(FR) SYSTÈME D'ESTIMATION DE MODÈLE, PROCÉDÉ D'ESTIMATION DE MODÈLE ET PROGRAMME D'ESTIMATION DE MODÈLE
(JA) モデル推定システム、モデル推定方法およびモデル推定プログラム
Abstract
(EN)
An input unit 81 accepts, as input thereto, action data in which an environment state and an action performed in said environment are correlated, a prediction model for predicting a state that corresponds to an action on the basis of the action data, and an explanatory variable to an objective function that evaluates the state and the action together. A structure setting unit 82 sets a branch structure in which an objective function is placed in the lowermost node of a hierarchical mixed expert model. A learning unit 83 learns the objective function that includes the explanatory variable and a branch condition in a node of the hierarchical mixed expert model on the basis of a state predicted by applying the prediction model to the action data, which is divided in accordance with the branch structure.
(FR)
L'invention concerne une unité d'entrée (81) qui reçoit en entrée des données d'action dans lesquelles un état d'environnement et une action effectuée dans ledit environnement sont corrélés, un modèle de prédiction permettant de prédire un état qui correspond à une action en fonction des données d'action, et une variable explicative à une fonction objective qui évalue l'état et l'action conjointement. Une unité de définition de structure (82) règle une structure de branche dans laquelle une fonction objective est placée dans le nœud le plus bas d'un modèle expert mixte hiérarchique. Une unité d'apprentissage (83) apprend la fonction objective qui comprend la variable explicative et une condition de branche dans un nœud du modèle d'expert mixte hiérarchique en fonction d'un état prédit par l'application du modèle de prédiction aux données d'action, qui sont divisées en fonction de la structure de branche.
(JA)
入力部81は、環境の状態とその環境の元で行われる行動とを対応付けたデータである行動データ、行動データに基づいて行動に応じた状態を予測する予測モデル、および、状態と行動とを合わせて評価する目的関数の説明変数とを入力する。構造設定部82は、階層混合エキスパートモデルの最下層のノードに目的関数が配される分岐構造を設定する。学習部83は、分岐構造に従って分割される行動データに対して予測モデルを適用して予測される状態に基づいて、階層混合エキスパートモデルのノードにおける分岐条件および説明変数を含む目的関数を学習する。
Also published as
JP2020508787
Latest bibliographic data on file with the International Bureau