Projected Latent Markov Chain Monte Carlo: Conditional Sampling of Normalizing Flows
arXiv:2007.06140
Abstract
We introduce Projected Latent Markov Chain Monte Carlo (PL-MCMC), a technique for sampling from the high-dimensional conditional distributions learned by a normalizing flow. We prove that a Metropolis-Hastings implementation of PL-MCMC asymptotically samples from the exact conditional distributions associated with a normalizing flow. As a conditional sampling method, PL-MCMC enables Monte Carlo Expectation Maximization (MC-EM) training of normalizing flows from incomplete data. Through experimental tests applying normalizing flows to missing data tasks for a variety of data sets, we demonstrate the efficacy of PL-MCMC for conditional sampling from normalizing flows.
arXiv comment27 pages, 22 figures, 4 tables
References in corpus (15)
- Conditional Generative Adversarial Nets
- MissForest - nonparametric missing value imputation for mixed-type data
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- WaveNet: A Generative Model for Raw Audio
- Conditional Image Generation with PixelCNN Decoders
- Pixel Recurrent Neural Networks
- NICE: Non-linear Independent Components Estimation
- The Perception-Distortion Tradeoff
- GAIN: Missing Data Imputation using Generative Adversarial Nets
- NeuTra-lizing Bad Geometry in Hamiltonian Monte Carlo Using Neural Transport
- Variational Autoencoder with Arbitrary Conditioning
- MisGAN: Learning from Incomplete Data with Generative Adversarial Networks
- Iterative Neural Autoregressive Distribution Estimator (NADE-k)
- Flow Models for Arbitrary Conditional Likelihoods
- On the convergence of the Metropolis-Hastings Markov chains