The effect of whitening on explanation performance

TL;DR

This work investigates whether data whitening can mitigate suppressor-driven misattributions in XAI explanations. By applying five whitening strategies to CORR-background data from the XAI-TRIS suite and evaluating 16 attribution methods across three model architectures, it shows that some whitening transforms (notably sphering, symmetric orthogonalization, and optimal signal preservation) can improve explanation fidelity, but improvements are method- and model-dependent and do not universally eliminate suppressors. A complementary 2D theoretical analysis reveals that, in linear settings, only Partial Regression can remove the suppressor influence, while other whitening methods fail to do so, with broader implications for high-dimensional, non-linear data. Overall, the results highlight the nuanced, non-universal role of preprocessing in explainability and suggest that preprocessing choices must be considered carefully when interpreting attributions in practice.

Abstract

Explainable Artificial Intelligence (XAI) aims to provide transparent insights into machine learning models, yet the reliability of many feature attribution methods remains a critical challenge. Prior research (Haufe et al., 2014; Wilming et al., 2022, 2023) has demonstrated that these methods often erroneously assign significant importance to non-informative variables, such as suppressor variables, leading to fundamental misinterpretations. Since statistical suppression is induced by feature dependencies, this study investigates whether data whitening, a common preprocessing technique for decorrelation, can mitigate such errors. Using the established XAI-TRIS benchmark (Clark et al., 2024b), which offers synthetic ground-truth data and quantitative measures of explanation correctness, we empirically evaluate 16 popular feature attribution methods applied in combination with 5 distinct whitening transforms. Additionally, we analyze a minimal linear two-dimensional classification problem (Wilming et al., 2023) to theoretically assess whether whitening can remove the impact of suppressor features from Bayes-optimal models. Our results indicate that, while specific whitening techniques can improve explanation performance, the degree of improvement varies substantially across XAI methods and model architectures. These findings highlight the complex relationship between data non-linearities, preprocessing quality, and attribution fidelity, underscoring the vital role of pre-processing techniques in enhancing model interpretability.

Figures (9)

• Figure 1: Average earth mover's distance (left) and precision (right) across all samples, XAI methods, and XAI-TRIS scenarios. This is split by background type, where the WHITE background serves as a 'baseline' with the CORR background type serving as the base for whitening. Each subsequent row therefore shows the application of different whitening methods to the underlying CORR background scenarios. Both metrics follow a similar trend of which whitening methods improve the correctness, where the sphering, symmetric orthogonalization, and optimal signal preserving methods perform the best -- nearly reaching the performance levels of the WHITE results.
• Figure 2: Data sampled from the linear generative process of the 2D-suppressor data wilming_theoretical_2023, with correlation coefficient $c=0.8$ and variances $s_1^2=1.0$ and $s_2^2=1.0$. Boundaries of the Bayes-optimal decision are shown as well, where the marginal distribution of suppressor variable $x_2$ for the original data shows that it does not carry any class-related information. Only Partial Regression is able to remove the influence of the suppressor from the Bayes-optimal decision.
• Figure 3: Absolute-valued global importance maps calculated as the mean importance value over all correctly predicted samples, for selected XAI methods and baselines.
• Figure 4: Absolute-valued importance maps obtained for a random correctly-predicted data sample, for data with no whitening applied and data for which the symmetric orthogonalization whitening method was applied. Note, different samples are visualised for both cases.
• Figure 5: Boxplots of EMD scores across all problem scenarios and background types, where each plot is separated for each of the four main XAI methods studied.