We propose to learn deep undirected graphical models (i.e., MRFs) with a non-ELBO objective for which we can calculate exact gradients. In particular, we optimize a saddle-point objective deriving from the Bethe free energy approximation to the partition function. Unlike much recent work in approximate inference, the derived objective requires no sampling, and can be efficiently computed even for very expressive MRFs. We furthermore amortize this optimization with trained inference networks. Experimentally, we find that the proposed approach compares favorably with loopy belief propagation, but is faster, and it allows for attaining better held out log likelihood than other recent approximate inference schemes.
Sam Wiseman (TTIC)
Yoon Kim (Harvard University)
More from the Same Authors
2018 Poster: Latent Alignment and Variational Attention »
Yuntian Deng · Yoon Kim · Justin Chiu · Demi Guo · Alexander Rush