Model agnostic local variable importance for locally dependent relationships
This work addresses the need for better interpretability in machine learning for researchers and practitioners dealing with complex, locally dependent data, though it appears incremental as it builds on existing local importance methods.
The paper tackled the problem of accurately assessing variable importance for individual observations in machine learning models, particularly for locally dependent relationships and multi-class classification, by proposing CLIQUE, a model-agnostic method that improves over permutation-based approaches and reduces bias in irrelevant regions.
Global variable importance measures are commonly used to interpret the results of machine learning models. Local variable importance techniques assess how variables contribute to individual observations. Current methods typically fail to accurately reflect locally dependent relationships between variables and instead focus on marginal importance values. Additionally, they are not natively adapted for multi-class classification problems. We propose a new model-agnostic method for calculating local variable importance, CLIQUE, that captures locally dependent relationships, improves over permutation-based methods, and can be directly applied to multi-category classification problems. Simulated and real-world examples show that CLIQUE emphasizes locally dependent information and properly reduces bias in regions where variables do not affect the response.