Probabilistic inference using contraction of tensor networks
TensorInference, a package for exact probabilistic inference in discrete graphical models, capitalizes on recent tensor network advancements. Its tensor-based engine features optimized contraction ordering methods, an aspect vital to computational performance. Additionally, it incorporates optimized BLAS routines and GPU technology for enhanced efficiency. In a comparative evaluation with similar libraries, TensorInference demonstrates superior scalability for models of increasing complexity.
Jul 10, 2024
JunctionTrees.jl - Efficient Bayesian Inference in Discrete Graphical Models
JunctionTrees.jl implements the junction tree algorithm, an efficient method to perform Bayesian inference in discrete probabilistic graphical models. It exploits Julia's metaprogramming capabilities to separate the algorithm into a compilation and a runtime phase. This opens a wide range of optimization possibilities in the compilation stage. The non-optimized runtime performance of JunctionTrees.jl is similar to those of analog C++ libraries such as libdai and Merlin.
Jul 27, 2022