방법 비교

	적대적 훈련	데이터 증강 (Data Augmentation)	생성적 적대 신경망
분야	딥러닝	딥러닝	딥러닝
계열	Machine learning	Machine learning	Machine learning
기원 연도	2018	2019	2014
창시자	Aleksander Madry et al.	Connor Shorten & Taghi Khoshgoftaar	Goodfellow, I. et al.
유형	Robust optimization training procedure	Regularization / data preprocessing technique	Generative deep learning (adversarial two-network game)
원전	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).	Shorten, C., & Khoshgoftaar, T. M. (2019). A survey on image data augmentation for deep learning. Journal of Big Data, 6, 60.	Goodfellow, I. et al. (2014). Generative Adversarial Nets. NeurIPS.
별칭	Min-Max Robust Training, PGD Adversarial Training, Robust Empirical Risk Minimization, Hasımsal Eğitim	Training Data Augmentation, Image Augmentation, Veri Artırma, Synthetic Data Augmentation	Üretici Çekişmeli Ağ (GAN), GAN, generative adversarial nets, adversarial network
관련	3	2	4
요약	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.	Data augmentation is a family of techniques that artificially expands a training dataset by applying label-preserving transformations to existing samples. Originally systematized for image classification tasks, it is now applied broadly across vision, text, audio, and tabular domains. It emerged as a practical answer to the chronic scarcity of labeled data in supervised deep learning and remains a standard preprocessing step in modern neural network pipelines.	A Generative Adversarial Network (GAN), introduced by Ian Goodfellow and colleagues in 2014, produces realistic synthetic data through the competition of two neural networks — a generator and a discriminator. It is widely used for image synthesis, data augmentation, and distribution estimation.