Differentiable samplers for deep latent variable models

• Published in: Philosophical transactions of the Royal Society of London. Series A: Mathematical, physical, and engineering sciences Vol. 381; no. 2247; p. 20220147
• Main Authors: Doucet, Arnaud, Moulines, Eric, Thin, Achille
• Format: Journal Article
• Language: English
• Published: England The Royal Society 15.05.2023
• Subjects: Review; Monte Carlo methods; importance sampling; Bayesian inference; variational inference

Latent variable models are a popular class of models in statistics. Combined with neural networks to improve their expressivity, the resulting deep latent variable models have also found numerous applications in machine learning. A drawback of these models is that their likelihood function is intrac...