Skip to yearly menu bar Skip to main content


MCMC Should Mix: Learning Energy-Based Model with Neural Transport Latent Space MCMC

Erik Nijkamp · Ruiqi Gao · Pavel Sountsov · Srinivas Vasudevan · Bo Pang · Song-Chun Zhu · Yingnian Wu

Keywords: [ mcmc ] [ generative models ] [ energy-based models ]


Learning energy-based model (EBM) requires MCMC sampling of the learned model as an inner loop of the learning algorithm. However, MCMC sampling of EBMs in high-dimensional data space is generally not mixing, because the energy function, which is usually parametrized by deep network, is highly multi-modal in the data space. This is a serious handicap for both theory and practice of EBMs. In this paper, we propose to learn EBM with a flow-based model (or in general latent variable model) serving as a backbone, so that the EBM is a correction or an exponential tilting of the flow-based model. We show that the model has a particularly simple form in the space of the latent variables of the generative model, and MCMC sampling of the EBM in the latent space mixes well and traverses modes in the data space. This enables proper sampling and learning of EBMs.

Chat is not available.