COT-GAN: Generating Sequential Data via Causal Optimal Transport

Abstract

We introduce COT-GAN, an adversarial algorithm to train implicit generative models optimized for producing sequential data. The loss function of this algorithm is formulated using ideas from Causal Optimal Transport (COT), which combines classic optimal transport methods with an additional temporal causality constraint. Remarkably, we find that this causality condition provides a natural framework to parameterize the cost function that is learned by the discriminator as a robust (worst-case) distance, and an ideal mechanism for learning time dependent data distributions. Following Genevay et al.\ (2018), we also include an entropic penalization term which allows for the use of the Sinkhorn algorithm when computing the optimal transport cost. Our experiments show effectiveness and stability of COT-GAN when generating both low- and high-dimensional time series data. The success of the algorithm also relies on a new, improved version of the Sinkhorn divergence which demonstrates less bias in learning.

References in corpus (15)

• GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
• Generating Videos with Scene Dynamics
• MMD GAN: Towards Deeper Understanding of Moment Matching Network
• C-RNN-GAN: Continuous recurrent neural networks with adversarial training
• Towards Principled Methods for Training Generative Adversarial Networks
• Real-valued (Medical) Time Series Generation with Recurrent Conditional GANs
• The Cramer Distance as a Solution to Biased Wasserstein Gradients
• Composing graphical models with neural networks for structured representations and fast inference
• Towards Accurate Generative Models of Video: A New Metric & Challenges
• Demystifying MMD GANs
• Sample Complexity of Sinkhorn divergences
• HYPE: A Benchmark for Human eYe Perceptual Evaluation of Generative Models
• From optimal transport to generative modeling: the VEGAN cookbook
• On gradient regularizers for MMD GANs
• Improving GANs Using Optimal Transport

Cited by in corpus (5)

• The Universal Approximation Property
• TCGAN: Convolutional Generative Adversarial Network for Time Series Classification and Clustering
• A Spatiotemporal Deep Neural Network for Fine-Grained Multi-Horizon Wind Prediction
• Towards Generating Real-World Time Series Data
• Generative Adversarial Network: Some Analytical Perspectives