Tensor networks contraction and the belief propagation algorithm
R. Alkabetz, Itai Arad
Abstract
The authors show how the belief-propagation algorithm can be used as an approximate tensor-network contraction algorithm, and that is equivalent to the simple-update method from the world of tensor networks.
Topics & Concepts
Belief propagationContraction (grammar)Graphical modelInferenceMathematicsTensor (intrinsic definition)Approximation algorithmApproximate inferenceAlgorithmSimple (philosophy)Applied mathematicsComputer sciencePure mathematicsArtificial intelligenceMedicineInternal medicineDecoding methodsStatisticsEpistemologyPhilosophyTensor decomposition and applicationsModel Reduction and Neural NetworksError Correcting Code Techniques