Explainability of Point Cloud Neural Networks Using SMILE: Statistical Model-Agnostic Interpretability with Local Explanations

TL;DR

This paper introduces SMILE, a statistical model-agnostic interpretability method for point cloud neural networks, enhancing explainability, robustness, and stability analysis, with implications for safety-critical autonomous systems.

Contribution

It adapts SMILE for point cloud models, incorporating ECDF distances for improved interpretability and robustness, and establishes a new benchmark for model stability using the Jaccard index.

Findings

SMILE outperforms existing methods in fidelity and robustness.

Introduces a stability benchmark for point cloud models.

Identifies dataset biases affecting classification accuracy.

Abstract

In today's world, the significance of explainable AI (XAI) is growing in robotics and point cloud applications, as the lack of transparency in decision-making can pose considerable safety risks, particularly in autonomous systems. As these technologies are integrated into real-world environments, ensuring that model decisions are interpretable and trustworthy is vital for operational reliability and safety assurance. This study explores the implementation of SMILE, a novel explainability method originally designed for deep neural networks, on point cloud-based models. SMILE builds on LIME by incorporating Empirical Cumulative Distribution Function (ECDF) statistical distances, offering enhanced robustness and interpretability, particularly when the Anderson-Darling distance is used. The approach demonstrates superior performance in terms of fidelity loss, R2 scores, and robustness across various kernel widths, perturbation numbers, and clustering configurations. Moreover, this study introduces a stability analysis for point cloud data using the Jaccard index, establishing a new benchmark and baseline for model stability in this field. The study further identifies dataset biases in the classification of the 'person' category, emphasizing the necessity for more comprehensive datasets in safety-critical applications like autonomous driving and robotics. The results underscore the potential of advanced explainability models and highlight areas for future research, including the application of alternative surrogate models and explainability techniques in point cloud data.