Current browse context:
cs
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: MCMC Should Mix: Learning Energy-Based Model with Neural Transport Latent Space MCMC
(Submitted on 12 Jun 2020 (v1), last revised 16 Mar 2022 (this version, v2))
Abstract: 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 a 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 an EBM with a flow-based model (or in general a 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 backbone 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.
Submission history
From: Erik Nijkamp [view email][v1] Fri, 12 Jun 2020 01:25:51 GMT (9068kb,D)
[v2] Wed, 16 Mar 2022 07:53:07 GMT (9340kb,D)
Link back to: arXiv, form interface, contact.