方法对比

	对抗训练	分布外检测
领域	深度学习	机器学习
方法族	Machine learning	Machine learning
起源年份	2018	2017
提出者	Aleksander Madry et al.	Hendrycks & Gimpel
类型	Robust optimization training procedure	Reliability and safety method for neural networks
开创性文献	Madry, A., Makelov, A., Schmidt, L., Tsipras, D., & Vladu, A. (2018). Towards deep learning models resistant to adversarial attacks. International Conference on Learning Representations (ICLR).	Hendrycks, D., & Gimpel, K. (2017). A baseline for detecting misclassified and out-of-distribution examples in neural networks. International Conference on Learning Representations.
别名	Min-Max Robust Training, PGD Adversarial Training, Robust Empirical Risk Minimization, Hasımsal Eğitim	OOD Detection, Novelty Detection, Open-Set Recognition, Dağılım Dışı Tespit
相关	3	3
摘要	Adversarial Training is a robust optimization procedure for deep neural networks in which the model is trained not on clean data alone but on worst-case perturbed inputs crafted during training. Formalized by Madry et al. (2018) as a min-max saddle-point problem, the method uses Projected Gradient Descent (PGD) to generate strong adversarial examples within a bounded Lp perturbation set before each gradient update, forcing the network to learn decision boundaries that are stable under such perturbations.	Out-of-Distribution (OOD) detection is a set of techniques that identify when a deployed machine learning model receives inputs that differ significantly from its training data distribution. Introduced as a formal problem by Hendrycks and Gimpel in 2017, these methods enable models to flag unfamiliar inputs rather than silently produce unreliable predictions, making them foundational to trustworthy and safe AI deployment in high-stakes domains.