Corpus ID: 232147810

Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models

@article{BondTaylor2021DeepGM,
  title={Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models},
  author={Sam Bond-Taylor and Adam Leach and Yang Long and Chris G. Willcocks},
  journal={ArXiv},
  year={2021},
  volume={abs/2103.04922}
}
Deep generative modelling is a class of techniques that train deep neural networks to model the distribution of training samples. Research has fragmented into various interconnected approaches, each of which making trade-offs including run-time, diversity, and architectural restrictions. In particular, this compendium covers energy-based models, variational autoencoders, generative adversarial networks, autoregressive models, normalizing flows, in addition to numerous hybrid approaches. These… Expand

References

SHOWING 1-10 OF 244 REFERENCES
Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
  • 3
  • Highly Influential
  • PDF
On the Anatomy of MCMC-based Maximum Likelihood Learning of Energy-Based Models
  • 43
  • Highly Influential
  • PDF
A RAD approach to deep mixture models
  • 20
  • Highly Influential
  • PDF
Large Scale GAN Training for High Fidelity Natural Image Synthesis
  • 1,694
  • Highly Influential
  • PDF
Glow: Generative Flow with Invertible 1x1 Convolutions
  • 948
  • Highly Influential
  • PDF
Spectral Normalization for Generative Adversarial Networks
  • 1,939
  • Highly Influential
  • PDF
Density estimation using Real NVP
  • 1,179
  • Highly Influential
  • PDF
Improved Variational Inference with Inverse Autoregressive Flow
  • 907
  • Highly Influential
  • PDF
Masked Autoregressive Flow for Density Estimation
  • 452
  • Highly Influential
  • PDF
Stacked Generative Adversarial Networks
  • 326
  • Highly Influential
  • PDF
...
1
2
3
4
5
...