Interpretable Two-level Boolean Rule Learning for Classification

Abstract

As a contribution to interpretable machine learning research, we develop a novel optimization framework for learning accurate and sparse two-level Boolean rules. We consider rules in both conjunctive normal form (AND-of-ORs) and disjunctive normal form (OR-of-ANDs). A principled objective function is proposed to trade classification accuracy and interpretability, where we use Hamming loss to characterize accuracy and sparsity to characterize interpretability. We propose efficient procedures to optimize these objectives based on linear programming (LP) relaxation, block coordinate descent, and alternating minimization. Experiments show that our new algorithms provide very good tradeoffs between accuracy and interpretability.

presented at 2016 ICML Workshop on Human Interpretability in Machine Learning (WHI 2016), New York, NY

References in corpus (2)

• Or's of And's for Interpretable Classification, with Application to Context-Aware Recommender Systems
• Learning Optimized Or's of And's

Cited by in corpus (9)

• A Survey on the Explainability of Supervised Machine Learning
• The Tsetlin Machine -- A Game Theoretic Bandit Driven Approach to Optimal Pattern Recognition with Propositional Logic
• Explainable Artificial Intelligence: a Systematic Review
• Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead
• Model Agnostic Multilevel Explanations
• TIP: Typifying the Interpretability of Procedures
• A Formal Framework to Characterize Interpretability of Procedures
• Learning Global Transparent Models Consistent with Local Contrastive Explanations
• Discovering Drug-Drug and Drug-Disease Interactions Inducing Acute Kidney Injury Using Deep Rule Forests