AI Prediction Confidence Interval Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current machine learning regression algorithms only produce mean vector predictions, failing to provide associated covariance matrices and confidence intervals, which are essential for risk-inclusive decision-making.
Innovation Solution
A system and method that generate both mean vector predictions and covariance matrices using historical data, determining predicted outcomes and confidence intervals by employing classifiers and binning techniques to calculate probabilities and relationships between input and output values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If machine learning regression algorithms produce only mean vector predictions, then the prediction process is simple and fast, but the output lacks covariance matrices and confidence intervals needed for risk-inclusive decision-making
Solution Approach 1:
The patent merges the mean vector prediction with the covariance matrix into a unified prediction output structure. The system combines these two components to provide complete statistical information about the prediction distribution, enabling risk-inclusive decision-making while maintaining a coherent prediction framework.
Solution Approach 2:
The prediction algorithm is designed to serve multiple functions simultaneously: it generates both the mean vector prediction and the covariance matrix from the same input data. This multi-functional approach ensures that the system provides complete statistical information (mean, variance, and their relationships) without requiring separate processing steps.
2Measurement precision
If the system generates classifiers and covariance matrices for each element of the prediction request, then prediction accuracy and reliability improve, but processing time and computational complexity increase
Solution Approach 1:
The patent segments the prediction request into individual elements, where each element is processed independently through its own classifier and binning process. This segmentation allows for parallel processing of multiple prediction elements, improving overall efficiency while maintaining the accuracy benefits of individualized classification for each element.
Solution Approach 2:
The system performs preliminary actions by pre-processing the training data into bins and pre-computing classifier models before actual prediction requests are processed. This preliminary preparation reduces the computational burden during real-time predictions, cutting processing time while preserving prediction accuracy through the pre-established classification frameworks.
Data Source
AI summary
A confidence interval determination system includes a processor and memory. The memory stores a historical data store and instructions. The instructions include, in response to receiving a prediction request, determining a size of each element of the prediction request and obtaining a previous prediction request. The instructions include, in response to the prediction request being different from the previous prediction request, for each element of the prediction request, generating a classifier based on the training data and, for each bin of a set of bins of the element, determining a set of predicted likelihoods using the classifier. The instructions include generating a mean vector prediction and a covariance matrix using the sets of bins generated, determining a predicted outcome and a confidence interval using the mean vector prediction and the covariance matrix, and, in response to the confidence interval exceeding a threshold, outputting the predicted outcome and the confidence interval.