在Kubernetes中用Kubeflow自动防御对抗攻击，保准确率
Enabling Adversarial Robustness in AI Models through Kubeflow MLOps
- 基于Kubeflow的MLOps架构，实时检测推理阶段攻击
- 检测到准确率下降后自动启用PGD防御，恢复准确率
- 适合云上部署模型的安全防护，尤其关注对抗攻击
AI模型正越来越多地部署在云原生环境中以支持可扩展和自动化的服务。然而，尽管Kubernetes等平台提供强大的基础设施编排能力，针对已部署AI模型的安全防护机制仍显不足。本文提出一种部署在Kubernetes集群中的AI模型安全方案。该架构基于Kubeflow MLOps，在推理阶段自动检测对抗攻击，并触发防御机制以维持模型的准确性和可靠性。具体而言，在推理时引入快速梯度符号法（FGSM）攻击，一旦检测到准确率下降，即自动部署基于投影梯度下降（PGD）的对抗训练防御。实验结果表明，所部署的防御有效提升了模型鲁棒性，显著恢复了因攻击导致的准确率下降。
原文摘要 · Abstract (English)
AI models are increasingly deployed in cloud-native environments to support scalable and automated services. However, while platforms such as Kubernetes provide strong infrastructure orchestration, security mechanisms specifically designed to protect deployed AI models remain limited. This paper presents security measures for AI models deployed in Kubernetes clusters. The proposed architecture integrates Kubeflow-based MLOps to automatically detect adversarial attacks during the inference phase and trigger defense mechanisms that preserve the model's accuracy and reliability. Specifically, a Fast Gradient Sign Method (FGSM) attack is applied at inference time, and a Projected Gradient Descent (PGD)-based adversarial training defense is automatically deployed when a degradation in accuracy is detected. The experimental results indicate that the deployed defense robustifies the model, significantly recovering accuracy relative to the degradation caused by the attack.
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。