Home / AIO Classes / Network / Bayesian Network

Bayesian Network

Network AIO:BayesianNetwork NOT_REVIEWED

A network that is a probabilistic graphical model representing variables and their conditional dependencies via a directed acyclic graph.

Ontology IRI: https://w3id.org/aio/BayesianNetwork

Ontology Structure & Relationships

Parent Classes

AIO:Network (Network)

Branches

Network

Annotations & Sources

Definition Sources:

https://en.wikipedia.org/wiki/Bayesian_network

Subsets:

https://w3id.org/aio/NetworkSubset

Curated AIMech Concepts Grounded to this Class

31 Concepts

Curated Concept	Kind	Mapping Relation	Rationale
Ancestral Gumbel-Top-k sampling `AIMech:AncestralGumbelTopKSampling`	Algorithms	related_match	The primary algorithm is explicitly defined for finite discrete Bayesian networks and uses their parent-conditioned probabilities to partition configuration space. The pinned AIO node denotes the directed acyclic probabilistic model being sampled. This is an algorithm-to-input-model related match, not equivalence, subclass membership or a claim that the sampler itself is a learned network.
Context-sensitive emission hidden Markov model `AIMech:ContextSensitiveEmissionHiddenMarkovModel`	Model Families	broad_match	For a fixed finite window, the hidden chain and observation children form a normalized directed acyclic factorization even when an observation has a later hidden-state parent. The AIO Bayesian-network family is broader than this emission-context sequence construction.
Deep autoregressive network `AIMech:DeepAutoregressiveNetwork`	Architectures	related_match	The decoder is a directed acyclic probabilistic model with ordered within-layer factors, matching the AIO target at the generative-model component level. The curated DARN identity also includes a separate recognition network and their paired architecture. A related match expresses this component relationship without equating the whole encoder/decoder system with a single Bayesian network or asserting an algorithm-to-network subclass.
Document NADE `AIMech:DocumentNADE`	Architectures	broad_match	For a given length, the observed-word variables have ordered directed conditional factors, giving a fully visible Bayesian network. The AIO target is broader than this shared additive-embedding and tree-output architecture. Deterministic topic features do not add stochastic latent nodes.
Factorial hidden Markov model `AIMech:FactorialHiddenMarkovModel`	Model Families	broad_match	For a finite sequence, the independent-chain model with shared observation children is a directed acyclic probabilistic network. The pinned BayesianNetwork definition has that broader scope and is parented under Network; the local factorial temporal structure is more specific. This is a local mapping, not an upstream axiom.
Fully visible sigmoid belief network `AIMech:FullyVisibleSigmoidBeliefNetwork`	Architectures	broad_match	The pinned AIO target is broader than this fully visible binary logistic family. Ordered visible variables and normalized forward conditional factors form a directed acyclic probabilistic network; the local identity adds logistic-linear parameterization and absence of latent random variables.
Helmholtz machine `AIMech:HelmholtzMachine`	Architectures	related_match	The original architecture includes a directed acyclic generative model and a distinct directed recognition model. The AIO target captures the directed probabilistic model component, whereas the native identity includes the paired-model arrangement and inference role. A related match avoids equating the entire pair with a single generative DAG or with an undirected recurrent network.
Hidden Markov decision tree `AIMech:HiddenMarkovDecisionTree`	Model Families	broad_match	The finite time-unrolled conditional factorization is a directed acyclic probabilistic network with inputs conditioned on. The pinned BayesianNetwork target is broader than this hierarchical temporal model.
Hidden Markov tree `AIMech:HiddenMarkovTree`	Model Families	broad_match	The hidden root, child transitions and site observation nodes have a normalized directed acyclic factorization. A rooted hidden tree with independent emission leaves is a restricted Bayesian-network family, so the AIO target is broader; this mapping adds no causal intervention interpretation.
Hidden semi-Markov model `AIMech:HiddenSemiMarkovModel`	Model Families	broad_match	The inspected finite-duration model has a normalized directed acyclic representation over duration-augmented hidden states and observation children. The Bayesian-network family is broader than this explicit-sojourn model.
Hierarchical mixture of experts `AIMech:HierarchicalMixtureOfExperts`	Model Families	broad_match	For a fixed input, latent decisions down the finite tree and a leaf-conditioned output define a normalized directed acyclic probabilistic factorization. The pinned Bayesian-network family is broader than this conditional hierarchical mixture.
INLA group node conflict diagnostic `AIMech:INLAGroupNodeConflictDiagnostic`	Algorithms	related_match	The diagnostic splits information about latent quantities in Bayesian hierarchical graphical models. BayesianNetwork provides representation context, not the INLA approximation or group-wise diagnostic; this is a related mapping, not an asserted subclass or equivalence.
Input-output hidden Markov model `AIMech:InputOutputHiddenMarkovModel`	Model Families	broad_match	Figure 1 and Section 2 explicitly define the model’s time-unrolled directed acyclic dependence structure, with inputs conditioned on. The AIO BayesianNetwork target is broader than this input-conditioned latent-state family.
Mixed memory Markov model `AIMech:MixedMemoryMarkovModel`	Model Families	broad_match	For a finite time window and specified initial history, normalized conditional factors and latent predecessor selectors form a directed acyclic probabilistic model. This compact mixed-memory parameterization is narrower than the pinned BayesianNetwork class.
Multiple-testing node conflict diagnostic `AIMech:MultipleTestingNodeConflictDiagnostic`	Algorithms	related_match	The algorithm forms joint conflict comparisons from separator copies in Bayesian directed acyclic evidence-synthesis graphs. BayesianNetwork supplies this representation context; it is not the diagnostic or its simultaneous testing procedure, so this is relatedness rather than equivalence or subclass membership.
Neural autoregressive distribution estimator `AIMech:NeuralAutoregressiveDistributionEstimator`	Architectures	broad_match	The original paper explicitly defines NADE as a fully visible Bayesian network: ordered visible variables form an acyclic dependency graph with normalized neural conditional factors. The pinned AIO target is broader than this shared-parameter autoregressive architecture. Deterministic hidden computations parameterize the factors and do not introduce latent random variables.
Neural variational inference and learning `AIMech:NeuralVariationalInferenceAndLearning`	Algorithms	related_match	The method is explicitly developed for directed latent-variable belief networks, using their joint probabilities in the learning objective. The pinned AIO node denotes a directed acyclic conditional-dependence model. This is a training-method-to-model related match, not equivalence or a claim that NVIL is itself a network architecture.
Noisy-OR belief network `AIMech:NoisyORBeliefNetwork`	Architectures	broad_match	The AIO target denotes the broader directed acyclic probabilistic network class. The native noisy-OR family narrows its binary conditional factors to the stated failure-product form; its explicitly qualified signed extension retains an acyclic normalized factor model only under its validity constraint. Neither noisy input layers nor causal graph terminology establishes this identity.
Order-agnostic NADE training `AIMech:OrderAgnosticNADETraining`	Algorithms	related_match	The procedure trains a family of fully visible autoregressive directed probabilistic models, each indexed by an ordering. The pinned AIO target identifies that model context. This is a training-method-to-model related match, not equivalence or subclass membership of an algorithm in the network class; a missing-input mask alone does not justify mapping to an AIO masking layer.
Pairwise Markov chain `AIMech:PairwiseMarkovChain`	Model Families	broad_match	For a finite horizon the normalized joint-state chain has an acyclic directed factorization. Splitting each joint kernel by the probability chain rule also gives an ordered within-slice factorization without imposing causal semantics.
Pairwise Markov tree `AIMech:PairwiseMarkovTree`	Model Families	broad_match	The joint paired-state law has a normalized directed acyclic factorization on a rooted tree, a restricted Bayesian-network topology. The mapping concerns paired stochastic states and imposes no causal interpretation or requirement that the hidden marginal preserve that graph.
Randomized latent score conflict diagnostic `AIMech:RandomizedLatentScoreConflictDiagnostic`	Algorithms	related_match	The method organizes Bayesian hierarchical information into parent and child components of a directed graphical model. BayesianNetwork is its representation context; it does not denote the latent score diagnostic, frozen-parent sampling or its calibration. This is a related mapping without equivalence or subclass assertion.
Real-valued neural autoregressive density estimator `AIMech:RealValuedNeuralAutoregressiveDensityEstimator`	Architectures	broad_match	The primary paper explicitly gives the ordered real-valued factorization as a fully visible Bayesian network. The AIO target is broader than this shared-parameter neural mixture-conditional architecture. Hidden neural activations parameterize factors and are not unobserved random variables. A time-series autoregression or language-model label does not define the same identity.
RNN-NADE `AIMech:RecurrentNeuralNetworkNeuralAutoregressiveDistributionEstimator`	Architectures	broad_match	The visible variables can be ordered first by frame time and then by within-frame position, with normalized neural conditional factors depending only on earlier visible variables. This is a fully visible acyclic probabilistic dependence model, so the pinned BayesianNetwork is broader. Deterministic recurrent and prefix computations parameterize factors and do not imply intervention-level causality.
Reweighted wake-sleep `AIMech:ReweightedWakeSleep`	Algorithms	related_match	The core procedure explicitly learns layered directed probabilistic models using joint evaluations and a recognition proposal. The pinned target is the directed model class, whereas RWS is a parameter-learning procedure with importance weighting. This related mapping does not imply equivalence, subclass membership, or applicability to every model without the required tractable probabilities.
Separator-node conflict diagnostic `AIMech:SeparatorNodeConflictDiagnostic`	Algorithms	related_match	The method operates on Bayesian DAG information partitions and replicated separator nodes. BayesianNetwork supplies its defining directed-graph representation but is not the diagnostic algorithm itself; this is relatedness, not equivalence or subclass membership. The general separator can lie outside a random-effects hierarchy.
Sigmoid belief network `AIMech:SigmoidBeliefNetwork`	Architectures	broad_match	The pinned AIO target is broader: a directed acyclic probabilistic conditional-dependence network. This native family additionally requires binary nodes and logistic weighted-parent forward conditionals. The mapping is from the narrower native architecture to the broader AIO class, not an equivalence or an assertion of causal semantics.
Structured variational inference for factorial HMMs `AIMech:StructuredVariationalInferenceForFactorialHMMs`	Algorithms	related_match	The procedure approximates hidden-state inference in a particular directed probabilistic sequence model. It is an algorithm applied to a Bayesian network, not itself the broader network class.
Triplet Markov chain `AIMech:TripletMarkovChain`	Model Families	broad_match	A finite normalized triplet-state chain has an acyclic directed factorization, with within-slice factors obtainable by the probability chain rule. The generic Bayesian-network family is broader and does not require the particular hidden/auxiliary/observed partition.
Variational message passing `AIMech:VariationalMessagePassing`	Algorithms	related_match	The original core algorithm performs approximate inference through the parent, child and coparent neighborhoods of a directed acyclic probabilistic model. The pinned BayesianNetwork class denotes that model architecture, whereas VMP denotes an inference procedure; this is a direct algorithm-to-model relation, not equivalence or an is-a assertion.
Wake-sleep algorithm `AIMech:WakeSleepAlgorithm`	Algorithms	related_match	The original procedure trains a layered directed generative model through its normalized conditional distributions and a separate recognition model. The pinned target denotes that directed probabilistic model class. This is a training-method-to-model related match, not a claim that the algorithm itself is a network or is equivalent to the AIO class.

Import Provenance & Upstream Synchronization

Source Repo: AIO GitHub

Release Version: aio.json

Commit: 397851c547…

License: CC-BY-4.0

Imported At: 2026-10-10T07:34:10Z

SHA256: 8f741d71408c…

Source File: data/aio/BayesianNetwork.yaml