Induction of Non-monotonic Logic Programs To Explain Statistical Learning Models
This work addresses the need for interpretable AI explanations in domains using ensemble tree models, though it is incremental as it builds on existing methods like SHAP and HUIM.
The paper tackles the problem of explaining statistical learning models by inducing non-monotonic logic programs, achieving significant improvements in classification metrics and running time compared to the state-of-the-art ALEPH system on UCI benchmarks.
We present a fast and scalable algorithm to induce non-monotonic logic programs from statistical learning models. We reduce the problem of search for best clauses to instances of the High-Utility Itemset Mining (HUIM) problem. In the HUIM problem, feature values and their importance are treated as transactions and utilities respectively. We make use of TreeExplainer, a fast and scalable implementation of the Explainable AI tool SHAP, to extract locally important features and their weights from ensemble tree models. Our experiments with UCI standard benchmarks suggest a significant improvement in terms of classification evaluation metrics and running time of the training algorithm compared to ALEPH, a state-of-the-art Inductive Logic Programming (ILP) system.