Training Generative Adversarial Networks by Solving Ordinary Differential Equations
This addresses training instability for GAN users, offering a novel approach that departs from existing methods, though it is incremental in improving stability.
The paper tackles the instability in Generative Adversarial Network (GAN) training by analyzing continuous-time dynamics and hypothesizing that discretization errors cause issues; it shows that using ODE solvers with a regularizer stabilizes training and outperforms baselines on CIFAR-10 and ImageNet.
The instability of Generative Adversarial Network (GAN) training has frequently been attributed to gradient descent. Consequently, recent methods have aimed to tailor the models and training procedures to stabilise the discrete updates. In contrast, we study the continuous-time dynamics induced by GAN training. Both theory and toy experiments suggest that these dynamics are in fact surprisingly stable. From this perspective, we hypothesise that instabilities in training GANs arise from the integration error in discretising the continuous dynamics. We experimentally verify that well-known ODE solvers (such as Runge-Kutta) can stabilise training - when combined with a regulariser that controls the integration error. Our approach represents a radical departure from previous methods which typically use adaptive optimisation and stabilisation techniques that constrain the functional space (e.g. Spectral Normalisation). Evaluation on CIFAR-10 and ImageNet shows that our method outperforms several strong baselines, demonstrating its efficacy.