Spatiotemporal Attacks for Embodied Agents

TL;DR

This paper introduces the first study of adversarial attacks on embodied agents, generating spatiotemporal perturbations that exploit interaction history to reveal vulnerabilities in dynamic environments.

Contribution

It proposes a novel method for creating 3D adversarial examples using spatiotemporal perturbations and a trajectory attention module for embodied agents.

Findings

Perturbations effectively attack embodied agents in dynamic scenes.

The method generalizes well across different tasks and settings.

Attacks reveal significant vulnerabilities in embodied agent models.

Abstract

Adversarial attacks are valuable for providing insights into the blind-spots of deep learning models and help improve their robustness. Existing work on adversarial attacks have mainly focused on static scenes; however, it remains unclear whether such attacks are effective against embodied agents, which could navigate and interact with a dynamic environment. In this work, we take the first step to study adversarial attacks for embodied agents. In particular, we generate spatiotemporal perturbations to form 3D adversarial examples, which exploit the interaction history in both the temporal and spatial dimensions. Regarding the temporal dimension, since agents make predictions based on historical observations, we develop a trajectory attention module to explore scene view contributions, which further help localize 3D objects appeared with the highest stimuli. By conciliating with clues from the temporal dimension, along the spatial dimension, we adversarially perturb the physical properties (e.g., texture and 3D shape) of the contextual objects that appeared in the most important scene views. Extensive experiments on the EQA-v1 dataset for several embodied tasks in both the white-box and black-box settings have been conducted, which demonstrate that our perturbations have strong attack and generalization abilities.