Coupled Machine Learning and Explainability Processes for Prediction Transparency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine-learning processes often operate as 'black boxes,' lacking transparency in the importance and impact of input features on their operations and outputs, which hinders organizations' ability to make informed decisions and predict customer attrition effectively.
Innovation Solution
The implementation of an apparatus and method that utilizes coupled machine-learning and explainability processes. This involves receiving interaction data, applying a first trained AI process to predict the likelihood of a target event, and a second process to generate explainability data characterizing the prediction, thereby providing transparency and actionable insights.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a machine-learning process is used to predict target events, then prediction accuracy is improved, but transparency and interpretability of the prediction process deteriorate
Solution Approach 1:
The patent introduces an explainability process as an intermediary between the machine-learning prediction process and the end user. This explainability process generates explanations that characterize the predicted likelihood by identifying important input features and their contributions, thereby mediating the loss of transparency while preserving prediction accuracy.
Solution Approach 2:
The system segments the overall prediction system into two distinct components: a machine-learning process responsible for prediction accuracy and an explainability process responsible for transparency. This segmentation allows each component to optimize its specific function without compromising the other.
2Device complexity
If a single machine-learning process is used, then device complexity is reduced, but the ability to provide both prediction and explanation deteriorates
Solution Approach 1:
The patent creates a unified system that performs multiple functions: the machine-learning process generates predictions while the coupled explainability process generates explanations. Together, they provide dual functionality (prediction and interpretation) within an integrated architecture that manages complexity.
Solution Approach 2:
The system merges the machine-learning process and the explainability process into a coupled system where both processes operate together. The explainability process receives the same input data and prediction output, combining prediction and explanation capabilities in an integrated manner.
Data Source
AI summary
The disclosed embodiments include computer-implemented systems and processes that train adaptively and deployment of coupled machine-learning and explainability processes within distributed computing environments. By way of example, an apparatus may receive first interaction data associated with a first temporal interval from a computing system. Based on an application of a first and a second trained artificial-intelligence process to an input dataset that includes at least a subset of the first interaction data, the apparatus may generate output data indicative of a predicted likelihood of an occurrence of a target event during a second temporal interval, and may generate explainability data that characterizes the predicted likelihood. The apparatus may also transmit portions the output and explainability data to the computing system, and the computing system may modify an operation of an executed application program in accordance with at least one the output or explainability data.