Abstract. Inference in expressive probabilistic models is generally intractable, which makes them diffi- cult to learn and limits their applicability. Sum-.

The Sum-Product Theorem: A Foundation for Learning Tractable Models. (Supplementary Material). Abram L. Friesen. AFRIESEN@CS.WASHINGTON.EDU. Pedro Domingos.

Abstract. Inference in expressive probabilistic models is generally intractable, which makes them diffi- cult to learn and limits their applicability. Sum-.

This paper generalizes the principle of summation to a much broader set of learning problems: all those where inference consists of summing a function over ...

Jun 19, 2016 · Sum-product networks are a class of deep models where, surprisingly, inference remains tractable even when an arbitrary number of hidden layers ...

Nov 11, 2016 · Sum-product networks are a class of deep models where, surprisingly, inference remains tractable even when an arbitrary number of hidden layers ...

... The Sum-Product Theorem: A Foundation for Learning Tractable Models Review ... Furthermore, a generic learning technique is proposed to learn the sum-product ...

Based on this, LearnSPF is able to learn tractable, high- treewidth models in any semiring. Specific choices for decomposition, clustering and leaf-creation ...