Modeling Structure with Undirected Neural Networks. (arXiv:2202.03760v2 [cs.LG] UPDATED)

Web: http://arxiv.org/abs/2202.03760

Neural networks are powerful function estimators, leading to their status as
a paradigm of choice for modeling structured data. However, unlike other
structured representations that emphasize the modularity of the problem --
e.g., factor graphs -- neural networks are usually monolithic mappings from
inputs to outputs, with a fixed computation order. This limitation prevents
them from capturing different directions of computation and interaction between
the modeled variables.

In this paper, we combine the representational strengths of factor graphs and
of …

arxiv lg modeling networks neural neural networks