Evolved Machine Learning Models Feature Transformation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Machine learning models are limited by their initial features and accuracy, as they rely on specific input data and do not evolve to improve predictions over time.
Innovation Solution
The development of an evolutionary machine learning model technique that generates new features by selecting and transforming important features from initial models, allowing the model to evolve and improve predictions through iterations, using techniques such as decision tree analysis and feature transformations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a machine learning model is trained with initial features, then the model can make predictions, but the prediction accuracy is limited by the specificity of the initial features
Solution Approach 1:
The patent applies the Dynamics principle by transforming the static feature set into a dynamic, evolving feature set. Initial features are automatically transformed into new features through iterative processes, allowing the model to adapt and improve prediction accuracy over time rather than being constrained by fixed initial features
Solution Approach 2:
The patent applies Preliminary action by pre-processing and transforming initial features before model training. The system automatically generates transformed features from initial features in advance, so that when the model trains, it already has access to enhanced feature representations that improve prediction accuracy
2Reliability
If the model structure is fixed after initial training, then the model is simple to deploy, but the model cannot improve predictions over time
Solution Approach 1:
The patent applies Self-service by enabling the machine learning model to automatically generate its own transformed features without external intervention. The system autonomously iterates through feature transformation, model training, and validation cycles, improving prediction reliability through self-evolution while managing complexity through automated processes
Solution Approach 2:
The patent applies Feedback by implementing iterative validation where model predictions are evaluated against validation data, and the results feed back into the feature transformation process. This closed-loop system continuously refines the model structure based on performance feedback, improving reliability while the automation manages complexity
Data Source
AI summary
A plurality of initial machine learning models are determined based on a plurality of original features. The plurality of initial machine learning models are filtered by selecting a subset of the initial machine learning models as one or more surviving machine learning models. One or more evolved machine learning models are generated. At least one of the evolved machine learning models is based at least in part on one or more new features, which are based at least in part on a transformation of at least one of features of the one or more surviving machine learning models. Corresponding validation scores associated with the one or more evolved machine learning models and corresponding validation scores associated with the one or more surviving machine learning models are compared. At least one of the one or more evolved machine learning models or the one or more surviving machine learning models are selected as one or more new selected surviving machine learning models.