MoE架构在语义分割中对抗攻击下更鲁棒，优于传统集成模型。
Towards Adversarial Robustness of Model-Level Mixture-of-Experts Architectures for Semantic Segmentation
- 用可学习门控机制动态加权专家模型输出，提升整体表现。
- 在城市与高速场景下，对单实例和通用白盒攻击更具抵抗力。
- 适合追求高鲁棒性的自动驾驶视觉系统研究者使用。
深度神经网络易受对抗攻击影响。更大模型通常更鲁棒，集成方法通过组合多个模型优势提升抗扰能力。传统集成采用确定性规则融合输出，而混合专家（MoE）引入可学习的门控组件，动态预测各专家输出权重，决定其对最终预测的贡献。尽管MoE在特定任务上表现优于集成模型，但其对抗脆弱性尚未被研究。本文评估了MoE在城市与高速公路交通场景语义分割中的对抗脆弱性。结果表明，在多数情况下，MoE对单实例及通用白盒攻击具有更强鲁棒性，并能更好地抵御迁移攻击。代码已开源。
原文摘要 · Abstract (English)
Vulnerability to adversarial attacks is a well-known deficiency of deep neural networks. Larger networks are generally more robust, and ensembling is one method to increase adversarial robustness: each model's weaknesses are compensated by the strengths of others. While an ensemble uses a deterministic rule to combine model outputs, a mixture of experts (MoE) includes an additional learnable gating component that predicts weights for the outputs of the expert models, thus determining their contributions to the final prediction. MoEs have been shown to outperform ensembles on specific tasks, yet their susceptibility to adversarial attacks has not been studied yet. In this work, we evaluate the adversarial vulnerability of MoEs for semantic segmentation of urban and highway traffic scenes. We show that MoEs are, in most cases, more robust to per-instance and universal white-box adversarial attacks and can better withstand transfer attacks. Our code is available at \url{https://github.com/KASTEL-MobilityLab/mixtures-of-experts/}.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。