Markov Chain Monte Carlo

Deep Generative Classifier with Short Run Inference

Deep generative classifier utilizes short-run Markov chain Monte Carlo inference, Langevin dynamics, and backpropagation through time to achieve similar classification accuracy to an analogous convolutional neural network, but with the added benefits that it may generate data, may learn unsupervised from additional unlabeled data, and it exhibits robustness to adversarial attacks, due to the stochasticity of the Langevin equation and the top-down architecture of the underlying generator network

Learning Multi-Layer Latent Variable Model via Variational Optimization of Short Run MCMC for Approximate Inference

Ran experiments for a short-run MCMC residual network that outperforms a variational autoencoder in terms of image reconstruction error and image synthesis quality, while not requiring the design of a separate inference network

Exact Sampling with Coupled Markov Chains and Swendsen-Wang Cluster Sampling of the Ising Model

Convergence analysis of exact sampling with Gibbs sampler and coupled Markov chains vs. cluster sampling with Swendsen-Wang algorithm