RHALE: Robust and Heterogeneity-aware Accumulated Local Effects
This work improves explainability in machine learning by addressing robustness and heterogeneity issues in feature effect estimation, though it is incremental as it builds on the existing ALE method.
The paper tackles limitations in the Accumulated Local Effects (ALE) explainability method by proposing RHALE, which quantifies heterogeneity in local effects and automatically determines optimal variable-size bins, demonstrating superiority on synthetic and real datasets with correlated features.
Accumulated Local Effects (ALE) is a widely-used explainability method for isolating the average effect of a feature on the output, because it handles cases with correlated features well. However, it has two limitations. First, it does not quantify the deviation of instance-level (local) effects from the average (global) effect, known as heterogeneity. Second, for estimating the average effect, it partitions the feature domain into user-defined, fixed-sized bins, where different bin sizes may lead to inconsistent ALE estimations. To address these limitations, we propose Robust and Heterogeneity-aware ALE (RHALE). RHALE quantifies the heterogeneity by considering the standard deviation of the local effects and automatically determines an optimal variable-size bin-splitting. In this paper, we prove that to achieve an unbiased approximation of the standard deviation of local effects within each bin, bin splitting must follow a set of sufficient conditions. Based on these conditions, we propose an algorithm that automatically determines the optimal partitioning, balancing the estimation bias and variance. Through evaluations on synthetic and real datasets, we demonstrate the superiority of RHALE compared to other methods, including the advantages of automatic bin splitting, especially in cases with correlated features.