Fast and Simple Explainability for Point Cloud Networks

TL;DR

This paper introduces a fast, simple, and scalable explainability method for point cloud neural networks, enabling real-time interpretability and robustness analysis with significantly reduced computational cost.

Contribution

It presents Feature Based Interpretability (FBI), a novel approach that computes pointwise importance efficiently, outperforming existing methods in speed and explainability for 3D point cloud models.

Findings

Achieves at least three orders of magnitude speedup over current XAI methods.

State-of-the-art results in classification explainability.

Effective in analyzing rotation invariance, robustness, and dataset bias.

Abstract

We propose a fast and simple explainable AI (XAI) method for point cloud data. It computes pointwise importance with respect to a trained network downstream task. This allows better understanding of the network properties, which is imperative for safety-critical applications. In addition to debugging and visualization, our low computational complexity facilitates online feedback to the network at inference. This can be used to reduce uncertainty and to increase robustness. In this work, we introduce \emph{Feature Based Interpretability} (FBI), where we compute the features' norm, per point, before the bottleneck. We analyze the use of gradients and post- and pre-bottleneck strategies, showing pre-bottleneck is preferred, in terms of smoothness and ranking. We obtain at least three orders of magnitude speedup, compared to current XAI methods, thus, scalable for big point clouds or large-scale architectures. Our approach achieves SOTA results, in terms of classification explainability. We demonstrate how the proposed measure is helpful in analyzing and characterizing various aspects of 3D learning, such as rotation invariance, robustness to out-of-distribution (OOD) outliers or domain shift and dataset bias.