Strudel: Learning Structured-Decomposable Probabilistic Circuits

by   Meihua Dang, et al.

Probabilistic circuits (PCs) represent a probability distribution as a computational graph. Enforcing structural properties on these graphs guarantees that several inference scenarios become tractable. Among these properties, structured decomposability is a particularly appealing one: it enables the efficient and exact computations of the probability of complex logical formulas, and can be used to reason about the expected output of certain predictive models under missing data. This paper proposes Strudel, a simple, fast and accurate learning algorithm for structured-decomposable PCs. Compared to prior work for learning structured-decomposable PCs, Strudel delivers more accurate single PC models in fewer iterations, and dramatically scales learning when building ensembles of PCs. It achieves this scalability by exploiting another structural property of PCs, called determinism, and by sharing the same computational graph across mixture components. We show these advantages on standard density estimation benchmarks and challenging inference scenarios.


page 1

page 2

page 3

page 4


Tractable Regularization of Probabilistic Circuits

Probabilistic Circuits (PCs) are a promising avenue for probabilistic mo...

Efficient and Reliable Probabilistic Interactive Learning with Structured Outputs

In this position paper, we study interactive learning for structured out...

A Compositional Atlas of Tractable Circuit Operations: From Simple Transformations to Complex Information-Theoretic Queries

Circuit representations are becoming the lingua franca to express and re...

Enforcing and Discovering Structure in Machine Learning

The world is structured in countless ways. It may be prudent to enforce ...

On Tractable Computation of Expected Predictions

Computing expected predictions has many interesting applications in area...

HyperSPNs: Compact and Expressive Probabilistic Circuits

Probabilistic circuits (PCs) are a family of generative models which all...

Einsum Networks: Fast and Scalable Learning of Tractable Probabilistic Circuits

Probabilistic circuits (PCs) are a promising avenue for probabilistic mo...