# On Unifying Deep Generative Models

arXiv: https://arxiv.org/abs/1706.00550 · HTML: https://mlsanity.com/paper/1706.00550/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Jun 2017.

## Built on by (most cited first, 15 of 15)

- [Unifying Diffusion Models' Latent Space, with Applications to CycleDiffusion and Guidance](https://mlsanity.com/paper/2210.05559/) (2022)
- [DeepSMOTE: Fusing Deep Learning and SMOTE for Imbalanced Data](https://mlsanity.com/paper/2105.02340/) (2021)
- [GATSBI: Generative Adversarial Training for Simulation-Based Inference](https://mlsanity.com/paper/2203.06481/) (2022)
- [Unifying Generative Models with GFlowNets and Beyond](https://mlsanity.com/paper/2209.02606/) (2022)
- [Unaligned Image-to-Image Translation by Learning to Reweight](https://mlsanity.com/paper/2109.11736/) (2021)
- [Unified Generation, Reconstruction, and Representation: Generalized Diffusion with Adaptive Latent Encoding-Decoding](https://mlsanity.com/paper/2402.19009/) (2024)
- [A Mathematical Framework for Learning Probability Distributions](https://mlsanity.com/paper/2212.11481/) (2022)
- [Inferential Wasserstein Generative Adversarial Networks](https://mlsanity.com/paper/2109.05652/) (2021)
- [Learning Fair Models without Sensitive Attributes: A Generative Approach](https://mlsanity.com/paper/2203.16413/) (2022)
- [Statistical Regeneration Guarantees of the Wasserstein Autoencoder with Latent Space Consistency](https://mlsanity.com/paper/2110.03995/) (2021)
- [A Deterministic Sampling Method via Maximum Mean Discrepancy Flow with Adaptive Kernel](https://mlsanity.com/paper/2111.10722/) (2021)
- [Error Bounds for a Matrix-Vector Product Approximation with Deep ReLU Neural Networks](https://mlsanity.com/paper/2111.12963/) (2021)
- [Approximating Constraint Manifolds Using Generative Models for Sampling-Based Constrained Motion Planning](https://mlsanity.com/paper/2204.06791/) (2022)
- [There Is No Standard Answer: Knowledge-Grounded Dialogue Generation with Adversarial Activated Multi-Reference Learning](https://mlsanity.com/paper/2210.12459/) (2022)
- [Clinically Plausible Pathology-Anatomy Disentanglement in Patient Brain MRI with Structured Variational Priors](https://mlsanity.com/paper/2211.07820/) (2022)

## Abstract (arXiv)

Deep generative models have achieved impressive success in recent years. Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs), as emerging families for generative model learning, have largely been considered as two distinct paradigms and received extensive independent studies respectively. This paper aims to establish formal connections between GANs and VAEs through a new formulation of them. We interpret sample generation in GANs as performing posterior inference, and show that GANs and VAEs involve minimizing KL divergences of respective posterior and inference distributions with opposite directions, extending the two learning phases of classic wake-sleep algorithm, respectively. The unified view provides a powerful tool to analyze a diverse set of existing model variants, and enables to transfer techniques across research lines in a principled way. For example, we apply the importance weighting method in VAE literatures for improved GAN learning, and enhance VAEs with an adversarial mechanism that leverages generated samples. Experiments show generality and effectiveness of the transferred techniques.