On Multilevel Monte Carlo Unbiased Gradient Estimation for Deep Latent Variable Models.
Yuyang ShiRob CornishPublished in: AISTATS (2021)
Keyphrases
- monte carlo
- gradient estimation
- latent variable models
- variance reduction
- latent variables
- importance sampling
- real valued
- hidden markov models
- markov chain
- latent dirichlet allocation
- temporal difference
- learning problems
- hidden variables
- particle filter
- probabilistic model
- prior knowledge
- posterior distribution
- worst case