# Dynamics-aware Adversarial Attack of 3D Sparse Convolution Network

Source: [Dynamics-aware Adversarial Attack of 3D Sparse Convolution Network](<https://www.sota2.com/research/paper/68fa6e671e209131cea3fc0d>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T00:56:59.119Z

Authors: An Tao, Yueqi Duan, He Wang, Ziyi Wu, Pengliang Ji, Haowen Sun, Jie Zhou, Jiwen Lu

Published: 2021-12-17T10:53:35.000Z

## Abstract

In this paper, we investigate the dynamics-aware adversarial attack problem in deep neural networks. Most existing adversarial attack algorithms are designed under a basic assumption -- the network architecture is fixed throughout the attack process. However, this assumption does not hold for many recently proposed networks, e.g. 3D sparse convolution network, which contains input-dependent execution to improve computational efficiency. It results in a serious issue of lagged gradient, making the learned attack at the current step ineffective due to the architecture changes afterward. To address this issue, we propose a Leaded Gradient Method (LGM) and show the significant effects of the lagged gradient. More specifically, we re-formulate the gradients to be aware of the potential dynamic changes of network architectures, so that the learned attack better "leads" the next step than the dynamics-unaware methods when network architecture changes dynamically. Extensive experiments on various datasets show that our LGM achieves impressive performance on semantic segmentation and classification. Compared with the dynamic-unaware methods, LGM achieves about 20% lower mIoU averagely on the ScanNet and S3DIS datasets. LGM also outperforms the recent point cloud attacks.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2112.09428>)

[PDF](<https://arxiv.org/pdf/2112.09428.pdf>)
