Exponential-family harmoniums (EFHs), which extend restricted Boltzmannmachines (RBMs) from Bernoulli random variables to other exponential families(Welling et al., 2005), are generative models that can be trained withunsupervised-learning techniques, like