AI Model Training Data Refinement via Synthetic Sampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning models trained with incomplete or poor-quality data sets produce inaccurate results, and there is a need for techniques to improve the accuracy of these models, especially when important ranges of user preferences are poorly represented, and to enhance the relevance of input variable data sets.
Innovation Solution
A computer-implemented method that includes applying under-sampling and over-sampling techniques to training data sets, generating updated input variables through frequency filtering, and automatically determining optimal hyperparameters for machine learning models, such as LightGBM and mixed effects random forests, to improve model training and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional machine learning training is used with incomplete or poor-quality data sets, then the training process is simpler, but the accuracy of the model results deteriorates
Solution Approach 1:
The system performs preliminary actions by automatically detecting data quality issues, identifying important ranges, and generating synthetic training data before the actual model training process. This includes detecting missing data within important ranges, determining which ranges are critical for the specific application, and pre-generating synthetic data to fill these gaps, thereby improving accuracy before training begins
Solution Approach 2:
The system introduces an intermediary layer between the raw incomplete data and the machine learning model. This intermediary automatically detects data quality issues, identifies important ranges through analysis, and generates synthetic training data that bridges the gap between poor-quality input data and the requirements for accurate model training
2Measurement precision
If automated techniques are implemented to improve data quality and model accuracy, then the accuracy of machine learning models improves, but the complexity of the system increases
Solution Approach 1:
The system performs self-service by automatically detecting its own data quality issues, identifying important ranges, and generating appropriate synthetic training data without requiring manual intervention. The system autonomously analyzes the training data, determines what is missing, and corrects the deficiencies automatically
Solution Approach 2:
The system changes parameters by dynamically adjusting the synthetic data generation process based on detected data quality issues and identified important ranges. It modifies the training data parameters (adding synthetic samples, adjusting distributions) to optimize model accuracy while adapting to the specific deficiencies in the input data
3Reliability
If synthetic training data is generated to fill important ranges, then the completeness of training data improves, but the processing time and computational resources increase
Solution Approach 1:
The system applies partial action by focusing synthetic data generation only on the specific important ranges that are missing data, rather than generating synthetic data for the entire dataset. It identifies and targets only the critical gaps in the training data distribution, thereby improving completeness efficiently without unnecessary processing overhead
4Measurement precision
If manual review and curation of training data is performed to ensure quality, then the quality of input variables improves, but the productivity and speed of model development decreases
Solution Approach 1:
The system performs self-service by automatically detecting data quality issues, identifying important ranges, and generating appropriate synthetic training data without requiring manual intervention. The system autonomously analyzes the training data, determines what is missing, and corrects the deficiencies automatically
Solution Approach 2:
The system replaces the mechanical process of manual data review and curation with an automated computational system. Instead of human experts manually examining and curating training data, the system uses automated algorithms to detect issues, identify important ranges, and generate synthetic data, thereby maintaining quality while dramatically improving productivity
Data Source
AI summary
A system includes memory hardware configured to store and processor hardware configured to execute instructions. The instructions include loading a machine learning model, loading a training data set, loading baseline hyperparameters, configuring the machine learning model with the baseline hyperparameters, providing the training data set as inputs to the machine learning model configured with the baseline hyperparameters to determine baseline performance metrics, determining whether the baseline performance metrics are above a threshold, saving the baseline hyperparameters as optimal hyperparameters in response to determining that the baseline performance metrics are above the threshold, configuring the machine learning model with the optimal hyperparameters, loading input variables, providing the input variables as inputs to the machine learning model configured with the optimal hyperparameters to generate output variables, saving the output variables to a database, and generating a graphical user interface configured to access and display the output variables.