2020/02/20 by Shangyuan Tong, Tong, Shangyuan, Timur Garipov +3
Computer Science · Mathematics · #Adversarial Robustness in Machine Learning #Adversarial system #Algorithm #Anomaly Detection Techniques and Applications #Artificial intelligence #Computer science #Construct (python library) #Convergence (economics) #Discriminator #FOS: Computer and information sciences #Generalizability theory #Generative Adversarial Networks and Image Synthesis #Generator (circuit theory) #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Machine learning #Mathematical optimization #Mathematics #Normalization (sociology) #Pairwise comparison #Power (physics) #Regularization (linguistics) #Statistics
paper · pdf · doi:10.48550/arxiv.2002.08621
openalex publication_date 2020/02/20 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Adversarial training methods typically align distributions by solving two-player games. However, in most current formulations, even if the generator aligns perfectly with data, a sub-optimal discriminator can still drive the two apart. Absent additional regularization, the instability can manifest itself as a never-ending game. In this paper, we introduce a family of objectives by leveraging pairwise discriminators, and show that only the generator needs to converge. The alignment, if achieved, would be preserved with any discriminator. We provide sufficient conditions for local convergence; characterize the capacity balance that should guide the discriminator and generator choices; and construct examples of minimally sufficient discriminators. Empirically, we illustrate the theory and the effectiveness of our approach on synthetic examples. Moreover, we show that practical methods derived from our approach can better generate higher-resolution images.