Model Update Necessity Determination via Deviation Calculation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing model update necessity determination systems for machine learning models used in predicting vehicle behavior struggle to accurately assess when updates are needed due to changes in situations within preset areas, leading to potential inaccuracies in prediction.
Innovation Solution
A system and method that calculates the number of deviations of target vehicle data from learning conditions to determine the necessity of updating the machine learning model, incorporating a number-of-deviations calculation unit and an update necessity determination unit to assess the need for model updates based on these deviations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the machine learning model is used to process information from a large number of target vehicles in a preset area, then the prediction capability is improved, but the model becomes inadequate when the situation within the area changes over time
Solution Approach 1:
The patent implements a dynamic update mechanism that automatically determines when the machine learning model needs to be updated by calculating the number of deviations between current target vehicle data and the learning conditions. This dynamic approach allows the system to adapt to changing situations within the preset area while maintaining prediction accuracy, resolving the contradiction between reliability and adaptability.
2Adaptability or versatility
If the machine learning model is updated frequently to adapt to changing situations, then the adaptability is improved, but the system complexity and computational resources increase
Solution Approach 1:
The patent changes the parameter for determining model updates from fixed time intervals or arbitrary triggers to a specific metric: the number of deviations between current data and learning conditions. This parameter-based approach provides a clear, quantifiable criterion for update necessity, reducing system complexity while maintaining adaptability.
Solution Approach 2:
The system implements a feedback mechanism where the number-of-deviations calculation unit continuously monitors the deviation between current target vehicle data and the learning conditions, and the update necessity determination unit uses this feedback to decide whether model updates are needed. This feedback loop enables adaptive updates without requiring complex external control systems.
3Measurement precision
If the number of deviations is calculated to determine update necessity, then the update timing accuracy is improved, but the computational load increases
Solution Approach 1:
The patent applies partial action by calculating deviations only for the specific parameters defined in the learning conditions rather than analyzing all possible data dimensions. This selective approach achieves sufficient update timing accuracy while minimizing the computational energy required for the deviation calculation process.
Data Source
AI summary
A model update necessity determination system is a model update necessity determination system that determines necessity of updating a machine learning model that performs learning using vehicle data under predetermined learning conditions and predicts changes in the behavior of a target vehicle based on target vehicle data acquired from target vehicles within a preset area. The system includes a number-of-deviations calculation unit that calculates the number of deviations in which the target vehicle data deviates from the learning condition based on the target vehicle data of the target vehicles in the area, and an update necessity determination unit that determines the necessity of updating the machine learning model.