Generative Flow Networks for Discrete Probabilistic Modeling. (arXiv:2202.01361v2 [cs.LG] UPDATED)

We present energy-based generative flow networks (EB-GFN), a novel
probabilistic modeling algorithm for high-dimensional discrete data. Building
upon the theory of generative flow networks (GFlowNets), we model the
generation process by a stochastic data construction policy and thus amortize
expensive MCMC exploration into a fixed number of actions sampled from a
GFlowNet. We show how GFlowNets can approximately perform large-block Gibbs
sampling to mix between modes. We propose a framework to jointly train a
GFlowNet with an energy function, so …

arxiv flow lg modeling networks probabilistic modeling