Loading Now

Summary of Commit: Certifying Robustness Of Multi-sensor Fusion Systems Against Semantic Attacks, by Zijian Huang et al.


COMMIT: Certifying Robustness of Multi-Sensor Fusion Systems against Semantic Attacks

by Zijian Huang, Wenda Chu, Linyi Li, Chejian Xu, Bo Li

First submitted to arxiv on: 4 Mar 2024

Categories

  • Main: Machine Learning (cs.LG)
  • Secondary: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)

     Abstract of paper      PDF of paper


GrooveSquid.com Paper Summaries

GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!

Summary difficulty Written by Summary
High Paper authors High Difficulty Summary
Read the original abstract here
Medium GrooveSquid.com (original content) Medium Difficulty Summary
The proposed COMMIT certify robustness of multi-sensor fusion systems against semantic attacks by providing a practical anisotropic noise mechanism that leverages randomized smoothing with multi-modal data, and efficient algorithms to compute the certification. The framework is evaluated on different settings using the CARLA simulation platform, showing that the certification for MSF models is at most 48.39% higher than that of single-modal models. This work aims to contribute an important step towards certifiably robust autonomous vehicles.
Low GrooveSquid.com (original content) Low Difficulty Summary
Multi-sensor fusion systems are crucial in autonomous vehicles to ensure their safety and reliability. However, these systems can be vulnerable to semantic attacks like rotation and shifting. Researchers have proposed empirical defenses but they can be attacked again by new adaptive attacks. To address this issue, a new framework called COMMIT is proposed that certifies the robustness of multi-sensor fusion systems against semantic attacks. The framework uses anisotropic noise mechanisms and efficient algorithms to compute the certification. This work aims to make autonomous vehicles safer and more reliable.

Keywords

* Artificial intelligence  * Multi modal