arXiv (Cornell University) · 2019 · 22 citations · 21 references
Natural SciencesMathematicsComputer VisionDiscrete MathematicsOptimizationContinuous OptimizationConvex OptimizationDeep LearningNeural NetworksAutoencodersContinuous MappingsConvergenceData DistributionOptimal TransportRegularity Theory
This work builds the connection between the regularity theory of optimal transportation map, Monge-Ampère equation and GANs, which gives a theoretic understanding of the major drawbacks of GANs: convergence difficulty and mode collapse. According to the regularity theory of Monge-Ampère equation, if the support of the target measure is disconnected or just non-convex, the optimal transportation mapping is discontinuous. General DNNs can only approximate continuous mappings. This intrinsic conflict leads to the convergence difficulty and mode collapse in GANs. We test our hypothesis that the supports of real data distribution are in general non-convex, therefore the discontinuity is unavoidable using an Autoencoder combined with discrete optimal transportation map (AE-OT framework) on the CelebA data set. The testing result is positive. Furthermore, we propose to approximate the continuous Brenier potential directly based on discrete Brenier theory to tackle mode collapse. Comparing with existing method, this method is more accurate and effective.
21
Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
Jun-Yan Zhu, Taesung Park, Phillip Isola et al. · arXiv (Cornell University) · 2017 · 2.3K citations · Full text
Sinkhorn Distances: Lightspeed Computation of Optimal Transport
Marco Cuturi · 2013 · 2.2K citations
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Tero Karras, Timo Aila, Samuli Laine et al. · arXiv (Cornell University) · 2017 · 1.6K citations · Full text
Improved Training of Wasserstein GANs
Ishaan Gulrajani, Faruk Ahmed, Martín Arjovsky et al. · arXiv (Cornell University) · 2017 · 1.5K citations · Full text