Information-theoretic Evolution of Model Agnostic Global Explanations
This work addresses the need for robust global explanations in AI interpretability, particularly for classification models in domains like digital marketing, though it appears incremental as it builds on existing local explanation methods.
The paper tackles the problem of explaining black-box machine learning models globally by developing a model-agnostic approach that uses an evolutionary algorithm with an information-theoretic fitness function to generate interpretable rules, showing it outperforms existing methods on various datasets and improves robustness to distributional shifts by incorporating out-of-distribution samples.
Explaining the behavior of black box machine learning models through human interpretable rules is an important research area. Recent work has focused on explaining model behavior locally i.e. for specific predictions as well as globally across the fields of vision, natural language, reinforcement learning and data science. We present a novel model-agnostic approach that derives rules to globally explain the behavior of classification models trained on numerical and/or categorical data. Our approach builds on top of existing local model explanation methods to extract conditions important for explaining model behavior for specific instances followed by an evolutionary algorithm that optimizes an information theory based fitness function to construct rules that explain global model behavior. We show how our approach outperforms existing approaches on a variety of datasets. Further, we introduce a parameter to evaluate the quality of interpretation under the scenario of distributional shift. This parameter evaluates how well the interpretation can predict model behavior for previously unseen data distributions. We show how existing approaches for interpreting models globally lack distributional robustness. Finally, we show how the quality of the interpretation can be improved under the scenario of distributional shift by adding out of distribution samples to the dataset used to learn the interpretation and thereby, increase robustness. All of the datasets used in our paper are open and publicly available. Our approach has been deployed in a leading digital marketing suite of products.