Information-theoretic partition-based models for interpretable machine learning

Yang, L.

Yang, L. (2024)

Information-theoretic partition-based models for interpretable machine learning

Doctoral Thesis

In this dissertation, we study partition-based models that can be used both for interpretable predictive modeling and for understanding data via interpretable patterns. Specifically, we study probabilistic rule-based models for multi-class classification and histogram models for discretization, explanatory data analysis, and conditional mutual information estimation.

Probabilistic Rules

Classification Rules

The Minimum Description Length Principle

Conditional Mutual Information Estimation

Explanatory Data Analysis

Multivariate Discretization

Interpretable Machine Learning