18067852. PRECOMPUTED EXPLANATION SCORES simplified abstract (International Business Machines Corporation)
Contents
PRECOMPUTED EXPLANATION SCORES
Organization Name
International Business Machines Corporation
Inventor(s)
Stefan A. G. Van Der Stockt of Austin TX (US)
ERIKA Agostinelli of Bristol (GB)
Edward James Biddle of Winchester (GB)
Sourav Mazumder of Contra Costa CA (US)
PRECOMPUTED EXPLANATION SCORES - A simplified explanation of the abstract
This abstract first appeared for US patent application 18067852 titled 'PRECOMPUTED EXPLANATION SCORES
Simplified Explanation:
The patent application describes a method, system, and computer program product for generating precomputed explanation scores in AI systems. This involves clustering labeled transactions, scoring their homogeneity, and generating explainability scores for clusters.
- Clustering labeled transactions based on input features
- Scoring homogeneity of clustered transactions based on output labels
- Selecting clusters based on homogeneity scoring
- Obtaining explainability scores for selected clusters
- Generating unified explainability scores for clusters
- Storing precomputed explanation scores
Key Features and Innovation:
- Utilizes labeled transactions to generate explainability scores
- Incorporates clustering and homogeneity scoring for better insights
- Provides precomputed explanations for AI systems
Potential Applications:
This technology can be applied in various industries such as finance, healthcare, and marketing for improving transparency and interpretability of AI systems.
Problems Solved:
This technology addresses the challenge of understanding and interpreting the decisions made by AI systems, especially in complex and critical applications.
Benefits:
- Enhances transparency and trust in AI systems
- Facilitates better decision-making based on explainable AI
- Improves accountability and compliance in regulated industries
Commercial Applications:
The technology can be used in financial institutions for risk assessment, in healthcare for diagnosis support, and in e-commerce for personalized recommendations, among other commercial applications.
Prior Art:
Readers can explore prior research on explainable AI, machine learning interpretability, and clustering techniques in the context of AI systems.
Frequently Updated Research:
Stay updated on the latest advancements in explainable AI, interpretability methods, and applications of precomputed explanation scores in AI systems.
Questions about AI:
1. How does this technology improve the transparency of AI systems? 2. What are the potential implications of using precomputed explanation scores in various industries?
Original Abstract Submitted
A method, system, and computer program product generate precomputed explanation scores in AI systems. The method includes obtaining a set of labeled transactions comprising input features and corresponding output labels generated by a machine learning (ML) model and generating an explainable artificial intelligence (XAI) module. The generating includes clustering the labeled transactions based on the input features, scoring homogeneity of the clustered transactions based on the corresponding output labels, and selecting at least one cluster from the clustered transactions based on the homogeneity scoring. The generating further includes obtaining, by an explainability model, explainability scores for transactions in the at least one cluster, generating a unified explainability score for the at least one cluster based on the explainability scores, and storing the unified explainability score in a set of precomputed explanations.
