Variational Approaches for Auto-Encoding Generative Adversarial Networks

2017/06/15 by Mihaela Rosca, Rosca, Mihaela, Balaji Lakshminarayanan +5 · 4 citations
Computer Science · Physics and Astronomy · #Digital Media Forensic Detection #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Model Reduction and Neural Networks

paper · pdf · doi:10.48550/arxiv.1706.04987

openalex publication_date 2017/06/15 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Auto-encoding generative adversarial networks (GANs) combine the standard GAN algorithm, which discriminates between real and model-generated data, with a reconstruction loss given by an auto-encoder. Such models aim to prevent mode collapse in the learned generative model by ensuring that it is grounded in all the available training data. In this paper, we develop a principle upon which auto-encoders can be combined with generative adversarial networks by exploiting the hierarchical structure of the generative model. The underlying principle shows that variational inference can be used a basic tool for learning, but with the in- tractable likelihood replaced by a synthetic likelihood, and the unknown posterior distribution replaced by an implicit distribution; both synthetic likelihoods and implicit posterior distributions can be learned using discriminators. This allows us to develop a natural fusion of variational auto-encoders and generative adversarial networks, combining the best of both these methods. We describe a unified objective for optimization, discuss the constraints needed to guide learning, connect to the wide range of existing work, and use a battery of tests to systematically and quantitatively assess the performance of our method.

Citations

• On distinguishability criteria for estimating generative models
• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Statistical inference for noisy nonlinear ecological dynamic systems
• Adversarial Feature Learning
• Learning in Implicit Generative Models
• Conditional Image Synthesis With Auxiliary Classifier GANs
• Mode Regularized Generative Adversarial Networks
• Improved Training of Wasserstein GANs
• BEGAN: Boundary Equilibrium Generative Adversarial Networks
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks
• Hierarchical Implicit Models and Likelihood-Free Variational Inference
• Adversarial Variational Bayes: Unifying Variational Autoencoders and Generative Adversarial Networks
• Learning Multiple Layers of Features from Tiny Images

Cited by

• Scalable Diffusion Transformer for Conditional 4D fMRI Synthesis
• Discovering Latent Graphs with GFlowNets for Diverse Conditional Image Generation
• Distribution Matching in Variational Inference
• DISSECT: Disentangled Simultaneous Explanations via Concept Traversals

Related

• On Unifying Deep Generative Models
• Adversarial Variational Bayes: Unifying Variational Autoencoders and Generative Adversarial Networks
• Implicit Discriminator in Variational Autoencoder
• "Best-of-Many-Samples" Distribution Matching
• Variational Inference using Implicit Distributions