AI Model Training Configuration for Self-Optimizing Communication Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face challenges in flexibly configuring information for model training to meet diverse and complex communication scenarios, particularly in self-optimizing systems where artificial intelligence models need to meet specific system indicators and requirements.
Innovation Solution
A method and device for obtaining and transmitting model training configuration information, including parameters, data source constraints, and performance requirements, to enable flexible configuration of AI models for various communication systems, allowing autonomous model training that meets system-specific needs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If artificial intelligence models are introduced to meet diverse communication system requirements, then system performance and adaptability are improved, but device complexity and configuration difficulty increase
Solution Approach 1:
The model training configuration information is segmented into multiple independent parameters including input parameter types, output parameter types, data source constraints, model complexity requirements, and performance requirements. This segmentation allows each aspect of model configuration to be independently managed and adjusted, reducing overall configuration complexity while maintaining adaptability to diverse communication scenarios
Solution Approach 2:
The patent creates a universal model training configuration framework that can be applied across different communication scenarios (D2D, M2M, V2V, V2X) and different AI model types. The standardized configuration parameters serve multiple functions by accommodating various communication requirements through a single unified configuration mechanism, thereby improving adaptability without proportionally increasing complexity
2Adaptability or versatility
If model training configuration information is obtained and transmitted between devices, then flexible model training is enabled, but information transmission overhead and processing time increase
Solution Approach 1:
The model training configuration information is prepared and determined in advance before actual model training begins. All necessary parameters including input/output types, data source constraints, and performance requirements are pre-configured, allowing the model training process to start immediately without delays for configuration decisions during training execution
Solution Approach 2:
The patent utilizes parameter-based configuration where all model training requirements are expressed as adjustable parameters. By changing parameter values rather than restructuring the entire configuration, the system achieves flexibility in adapting to different scenarios while maintaining efficient processing through standardized parameter manipulation rather than complex reconfiguration
Data Source
Figure 1~2
Figure 3~4
Figure 5~8
AI summary
The present application relates to an information configuration method. The method comprises: a first device acquiring model training configuration information, wherein the model training configuration information is used for indicating information related to the training of a first model. In the embodiments of the present application, model training configuration information is designed according to actually required model characteristics, a model is trained according to the model training configuration information, a model training report is generated, and a model identifier is requested for a generated target model, such that the automated management of model training is realized.