Calibrating Energy-based Generative Adversarial Networks

Abstract

In this paper, we propose to equip Generative Adversarial Networks with the ability to produce direct energy estimates for samples.Specifically, we propose a flexible adversarial training framework, and prove this framework not only ensures the generator converges to the true data distribution, but also enables the discriminator to retain the density information at the global optimal. We derive the analytic form of the induced solution, and analyze the properties. In order to make the proposed framework trainable in practice, we introduce two effective approximation techniques. Empirically, the experiment results closely match our theoretical analysis, verifying the discriminator is able to recover the energy of data distribution.

ICLR 2017 camera ready

Cited by in corpus (12)

• On the Origin of Deep Learning
• Stacked Generative Adversarial Networks
• Augmented Normalizing Flows: Bridging the Gap Between Generative Flows and Latent Variable Models
• Exponential Family Estimation via Adversarial Dynamics Embedding
• Boltzmann machines and energy-based models
• Bounds all around: training energy-based models with bidirectional bounds
• Learning Latent Space Energy-Based Prior Model
• Robust GANs against Dishonest Adversaries
• Generative Modeling by Inclusive Neural Random Fields with Applications in Image Generation and Anomaly Detection
• Bridging Explicit and Implicit Deep Generative Models via Neural Stein Estimators
• Representation Learning: A Statistical Perspective
• Conditional Transferring Features: Scaling GANs to Thousands of Classes with 30% Less High-quality Data for Training