An introduction to deep generative modeling
From MaRDI portal
Abstract: Deep generative models (DGM) are neural networks with many hidden layers trained to approximate complicated, high-dimensional probability distributions using a large number of samples. When trained successfully, we can use the DGMs to estimate the likelihood of each observation and to create new samples from the underlying distribution. Developing DGMs has become one of the most hotly researched fields in artificial intelligence in recent years. The literature on DGMs has become vast and is growing rapidly. Some advances have even reached the public sphere, for example, the recent successes in generating realistic-looking images, voices, or movies; so-called deep fakes. Despite these successes, several mathematical and practical issues limit the broader use of DGMs: given a specific dataset, it remains challenging to design and train a DGM and even more challenging to find out why a particular model is or is not effective. To help advance the theoretical understanding of DGMs, we introduce DGMs and provide a concise mathematical framework for modeling the three most popular approaches: normalizing flows (NF), variational autoencoders (VAE), and generative adversarial networks (GAN). We illustrate the advantages and disadvantages of these basic approaches using numerical experiments. Our goal is to enable and motivate the reader to contribute to this proliferating research area. Our presentation also emphasizes relations between generative modeling and optimal transport.
Recommendations
Cites work
- A computational fluid mechanics solution to the Monge-Kantorovich mass transfer problem
- A multilevel method for the solution of time dependent optimal transport
- An Introduction to Variational Autoencoders
- Deep learning
- Deep learning: an introduction for applied mathematicians
- scientific article; zbMATH DE number 1909499 (Why is no real title available?)
- scientific article; zbMATH DE number 1444745 (Why is no real title available?)
- Optimization methods for large-scale machine learning
- Scikit-learn: machine learning in Python
- Stabilizing Invertible Neural Networks Using Mixture Models
Cited in
(12)- scientific article; zbMATH DE number 7027884 (Why is no real title available?)
- Generalized Normalizing Flows via Markov Chains
- Generative diffusion in very large dimensions
- State-observation augmented diffusion model for nonlinear assimilation with unknown dynamics
- Neural triangular map for density estimation and sampling with application to Bayesian inference
- Nonlinear denoising score matching for enhanced learning of structured distributions
- Efficient neural network approaches for conditional optimal transport with applications in Bayesian inference
- Generative assignment flows for representing and learning joint distributions of discrete data
- Diffusion map particle systems for generative modeling
- Data-driven methods for quantitative imaging
- Arbitrary distributions mapping via SyMOT-flow: a flow-based approach integrating maximum mean discrepancy and optimal transport
- KL convergence guarantees for score diffusion models under minimal data assumptions
This page was built for publication: An introduction to deep generative modeling
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6068234)
