Genetic Fuzzy Tree for Dynamic Parameter Tuning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Standard Genetic Algorithms (GAs) face scalability issues when dealing with complex problems that have numerous inputs and outputs, leading to computationally intractable solutions, especially in artificial intelligence applications where traditional methods would require excessively long training times.
Innovation Solution
The Genetic Fuzzy Tree (GFT) system employs a custom, highly optimized GA with dynamic parameter tuning, recursive learning capabilities, and parallel processing to efficiently manage complex problems, allowing for the creation and optimization of Fuzzy Inference Systems (FIS) that can handle hundreds or thousands of inputs and outputs within a significantly reduced timeframe.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If standard Genetic Algorithms are used to solve complex AI problems with numerous inputs and outputs, then the system can handle complex problems, but the computational time becomes excessively long and the solution becomes computationally intractable
Solution Approach 1:
The patent segments the complex problem-solving process into multiple hierarchical levels. The Genetic Fuzzy Tree divides the solution space into manageable segments by creating a tree structure where different parts of the chromosome handle different aspects of the problem, allowing parallel evaluation and reducing overall computational time
Solution Approach 2:
The patent implements nested structures where Fuzzy Inference Systems are nested within the Genetic Algorithm framework, and multiple FIS are organized in tree structures. This nesting allows the system to handle complex problems by breaking them down into smaller sub-problems that can be evaluated more efficiently at each level of the hierarchy
2Measurement precision
If the complexity of each chromosome or organism becomes quite high to solve complex problems, then the solution quality improves, but GAs face scalability issues
Solution Approach 1:
The chromosome is segmented into multiple genes, each representing a Fuzzy Inference System or a component thereof. This segmentation allows the complex chromosome to be evaluated in parts, improving scalability while maintaining high solution quality through the coordinated action of multiple specialized genes
Solution Approach 2:
Different parts of the chromosome (genes) are assigned different levels of complexity and specialization based on the specific requirements of different problem aspects. This local quality approach allows high complexity where needed while maintaining lower complexity elsewhere, improving overall scalability
Data Source
AI summary
The present disclosure describes a system and method for improving the way computing devices execute genetic algorithms. A fuzzy logic controller takes various properties of the genetic algorithm (such as the diversity of the population, the performance history of the algorithm in terms of time-efficiency and/or effectiveness at improving the best fitness function results, and available computing resources) to dynamically manage the parameters of the genetic algorithm. In some embodiments, the fuzzy inference system that provides parameters to the genetic algorithm is itself controlled by another fuzzy inference system.