Genetic Algorithm Model Architecture Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing model generation techniques are limited in flexibility, particularly in configuring layers and using input data types effectively, leading to suboptimal model performance.
Innovation Solution
An information processing method using a genetic algorithm to select and combine data types for input to model blocks, optimizing the model generation process by iteratively refining the selection of input data types and model architecture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a model with fixed layer configuration is generated using conventional techniques, then the model generation process is simple, but the model cannot flexibly use input data of different types
Solution Approach 1:
The patent applies dynamics by making the model architecture adaptable and changeable rather than fixed. The system dynamically selects and configures layers based on input data types, allowing the model structure to evolve and adapt to different data scenarios, thereby achieving flexibility without permanent complexity
Solution Approach 2:
The patent changes parameters by optimizing model architecture parameters (layer types, connections, configurations) based on the specific input data types. This allows the model to adjust its structural parameters to match the data characteristics, improving adaptability while managing complexity through parameter optimization
2Manufacturing precision
If layers are connected in series with fixed configuration, then the model structure is simple to generate, but the model performance is suboptimal for different data types
Solution Approach 1:
The patent applies preliminary action by pre-defining a library of available layer types and configurations that can be selected during model generation. This preparation work enables the system to quickly assemble optimized architectures for different data types without complex real-time decisions, improving both accuracy and generation efficiency
Solution Approach 2:
The patent optimizes model accuracy by adjusting architectural parameters such as layer types, connection patterns, and data type mappings based on the specific input data characteristics. This parameter optimization allows the model to achieve high precision for different data types while managing complexity through systematic parameter selection
Data Source
AI summary
An information processing method according to the present application is an information processing method executed by a computer, the information processing method including: acquiring learning data used for learning of a model having at least one block to which an output from an input layer is input, the learning data including a plurality of types of information; and selecting a type included in data to be input to the block by processing based on a genetic algorithm in learning using the learning data, and generating the model by using data corresponding to a combination of types selected among the plurality of types as an input from the input layer to the block.