Dual-Level Machine Control for Adaptive AI Input Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current artificial intelligence systems, particularly neural networks, are specialized for specific tasks and become ineffective when applied to different areas, requiring retraining and resulting in increased training and test data needs, making real-time responses impractical due to high dimensionality and complexity.
Innovation Solution
A method involving a control system with a working level and an evaluation level, where input values are processed to determine output values, with the evaluation level influencing the working level, allowing for adaptive processing and decision-making that considers both technical and non-technical conditions, such as moral-ethical aspects, using coupled artificial learning units and modulation functions to adjust network parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If specialized neural networks are used for specific tasks, then task performance is improved, but adaptability to different applications deteriorates
Solution Approach 1:
The patent implements a dual-level architecture where the working level handles specific task processing while the evaluation level provides cross-application assessment. This allows the system to maintain specialized processing capabilities for different tasks while achieving universality through the shared evaluation framework that can assess multiple application domains.
Solution Approach 2:
The system segments AI functionality into two distinct levels: the working level for task-specific operations and the evaluation level for comprehensive assessment. This segmentation allows each level to be optimized independently - the working level for specific task performance and the evaluation level for adaptability across applications.
2Measurement precision
If neural networks are retrained for different applications, then task specificity is improved, but training time and data requirements increase
Solution Approach 1:
The evaluation level is pre-configured with assessment capabilities for multiple applications, performing preliminary preparation work. This allows the system to quickly adapt to new applications by leveraging the pre-established evaluation framework rather than retraining entire networks from scratch.
Solution Approach 2:
The evaluation level acts as an intermediary between different applications and the working level. It provides a translation and adaptation layer that enables task-specific networks to serve multiple applications without requiring retraining, by mediating between the specialized working level and diverse application requirements.
3Speed
If high-dimensional data is processed in real-time, then response speed is improved, but system complexity increases
Solution Approach 1:
The system segments the processing of high-dimensional data across two levels: the working level performs initial rapid processing for immediate responses, while the evaluation level handles more complex comprehensive assessment. This segmentation enables real-time response for critical operations while distributing complex processing tasks.
Solution Approach 2:
The working level performs partial processing of high-dimensional data to achieve real-time responses for critical decisions, while the evaluation level performs excessive or complete processing when time permits. This allows the system to balance response speed with processing completeness based on situational requirements.
Data Source
AI summary
The invention relates to a method implemented in a control system of a machine for processing input values in an overall system with a working level and an evaluation level comprisinginputting first input values to the working level and determining first output values; forming first situation data; inputting the first situation data to the evaluation level and determining first evaluations indicating whether the first situation data satisfy predetermined first conditions; influencing the determination of the first output values based on the first evaluations; inputting second input values to the working level and determining second output values, wherein the determination of the second output values is influenced by the first output values; forming second situation data; inputting the second situation data to the evaluation level and determining second evaluations indicating whether the second situation data satisfy predetermined second conditions, the determining of the second evaluations being influenced by the first evaluations; influencing the determining of the second output values based on the second evaluations; wherein the first and/or the second output values are used as overall output values of the overall system.