Machine learning device, inference device, machine learning method, recording medium, and method for generating trained model
By training individual models for each stage of a problem and integrating them with an overall model using loss-based updates, the method enhances learning efficiency and accuracy in complex scenarios like tank base operations, addressing the balance challenge in existing methods.
WO2026140532A1PCT designated stage Publication Date: 2026-07-02ENEOS HLDG INC
Patent Information
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- ENEOS HLDG INC
- Filing Date
- 2025-11-07
- Publication Date
- 2026-07-02
Smart Images
Figure JP2025039069_02072026_PF_FP_ABST
Abstract
Provided is a machine learning device that, for a problem including a plurality of stages, performs training on a plurality of individual models (MP) for outputting an intermediate result for the problem, the individual models (MP) respectively corresponding to the stages, and an overall model (MS) that receives the output of each individual model (MP) as an input and outputs the final result for the problem. The machine learning device comprises a memory that stores a program, and a processor that executes the program stored in the memory. The processor acquires information outputted by each individual model (MP) as first information, calculates a loss corresponding to each piece of the first information as a first loss on the basis of a loss function corresponding to each of the individual models (MP), and updates the overall model (MS) on the basis of the first loss.
Need to check novelty before this filing date? Find Prior Art