Restricted Boltzmann Machine
Network AIO:RestrictedBoltzmannMachine NOT_REVIEWED
A Boltzmann machine network that learns the probability distribution of its input data.
Ontology IRI:
https://w3id.org/aio/RestrictedBoltzmannMachine
Ontology Structure & Relationships
Parent Classes
Branches
Annotations & Sources
Synonyms:
- RBM exact
Definition Sources:
Subsets:
https://w3id.org/aio/NetworkSubset
Curated AIMech Concepts Grounded to this Class
20 Concepts| Curated Concept | Kind | Mapping Relation | Rationale |
|---|---|---|---|
Autoregressive conditional restricted Boltzmann machine
AIMech:AutoregressiveConditionalRestrictedBoltzmannMachine |
Architectures | related_match | Each time step uses a conditional RBM with undirected visible–hidden interactions. The entire sequence construction additionally has directed history connections and a separate conditional normalizer per step; a related mapping records the component relationship without equating the whole temporal architecture to one unconditioned symmetric RBM. |
CD-PercLoss
AIMech:CDPercLoss |
Algorithms | related_match | This is a learning and prediction procedure developed for conditional RBM free energies. The AIO target names the model component/family, not this procedure; the association is related rather than an algorithm-is-model subclass assertion. |
Conditional restricted Boltzmann machine
AIMech:ConditionalRestrictedBoltzmannMachine |
Architectures | related_match | The construction uses a conditional RBM as its output distribution. This related mapping identifies that component or conditioned-family relationship without asserting that the whole input-conditioned or recurrent model is one unconditioned symmetric RBM. |
Contrastive divergence
AIMech:ContrastiveDivergence |
Algorithms | related_match | The inspected algorithm trains RBM parameters, but is not itself the model architecture. CD also applies beyond RBMs. This is a training-method to model-family association, not equivalence or subclass identity. |
Exponential-family harmonium
AIMech:ExponentialFamilyHarmonium |
Model Families | related_match | The source generalizes binary RBM/harmonium structure to exponential-family variables. The pinned RBM definition does not resolve the full general sufficient-statistic and normalization domain, so neither exact identity nor an unsupported subclass direction is asserted. |
Factored restricted Boltzmann machine language model
AIMech:FactoredRestrictedBoltzmannMachineLanguageModel |
Architectures | related_match | The architecture uses an undirected RBM interaction between the current categorical word and binary hidden state, with context affecting hidden inputs. A related mapping records that explicit component/conditional construction without asserting that the entire autoregressive or temporally linked sequence model is one unconditioned RBM. |
Family-of-harmonium
AIMech:FamilyOfHarmonium |
Model Families | related_match | Bipartite conditional-independence structure is shared with harmoniums, but the pinned RBM class does not identify this exact observation, hidden-support and parameter-sharing construction. |
Fast persistent contrastive divergence
AIMech:FastPersistentContrastiveDivergence |
Algorithms | related_match | The inspected algorithm trains RBM parameters, but is not itself the model architecture. CD also applies beyond RBMs. This is a training-method to model-family association, not equivalence or subclass identity. |
Generalized component analysis
AIMech:GeneralizedComponentAnalysis |
Model Families | related_match | GCA has the bipartite conditional-independence structure of an exponential-family harmonium, but its lognormal/Gaussian hidden law and heterogeneous categorical positions are not asserted identical to the underspecified pinned RBM class. |
HashCRBM
AIMech:HashCRBM |
Algorithms | related_match | This is a learning and prediction procedure developed for conditional RBM free energies. The AIO target names the model component/family, not this procedure; the association is related rather than an algorithm-is-model subclass assertion. |
Hierarchical harmonium
AIMech:HierarchicalHarmonium |
Model Families | related_match | Bipartite conditional-independence structure is shared with harmoniums, but the pinned RBM class does not identify this exact observation, hidden-support and parameter-sharing construction. |
Persistent contrastive divergence
AIMech:PersistentContrastiveDivergence |
Algorithms | related_match | The inspected algorithm trains RBM parameters, but is not itself the model architecture. CD also applies beyond RBMs. This is a training-method to model-family association, not equivalence or subclass identity. |
Product of experts
AIMech:ProductOfExperts |
Model Families | narrow_match | The local product-model family is broader than the AIO RBM target: the primary source explicitly derives RBMs as a product with one expert per hidden unit, while other expert families include HMMs. Narrow match means the target is narrower; this is a local scope assessment, not a new upstream axiom. FactorizedLLM’s related synonym alone does not establish equivalent generic scope. |
Rate-adapting Poisson model
AIMech:RateAdaptingPoissonModel |
Model Families | related_match | Bipartite conditional-independence structure is shared with harmoniums, but the pinned RBM class does not identify this exact observation, hidden-support and parameter-sharing construction. |
RNN-RBM
AIMech:RecurrentNeuralNetworkRestrictedBoltzmannMachine |
Architectures | related_match | The construction uses a conditional RBM as its output distribution. This related mapping identifies that component or conditioned-family relationship without asserting that the whole input-conditioned or recurrent model is one unconditioned symmetric RBM. |
Recurrent temporal restricted Boltzmann machine
AIMech:RecurrentTemporalRestrictedBoltzmannMachine |
Architectures | related_match | Each time step uses a conditional RBM with undirected visible–hidden interactions. The entire sequence construction additionally has directed history connections and a separate conditional normalizer per step; a related mapping records the component relationship without equating the whole temporal architecture to one unconditioned symmetric RBM. |
Replicated Softmax
AIMech:ReplicatedSoftmax |
Architectures | broad_match | The inspected local model is explicitly an RBM with categorical visible units, shared word parameters and length-scaled hidden biases. The pinned AIO RBM class is broader and does not specify those document constraints. |
Restricted Boltzmann Machine Collaborative Filtering
AIMech:RestrictedBoltzmannMachineCollaborativeFiltering |
Model Families | related_match | The construction uses shared RBM rating distributions, including mask-conditioned variants. A related mapping preserves that component/family relationship without equating the complete mask-indexed construction to one unconditional pinned RBM. |
Temporal factored restricted Boltzmann machine language model
AIMech:TemporalFactoredRestrictedBoltzmannMachineLanguageModel |
Architectures | related_match | The architecture uses an undirected RBM interaction between the current categorical word and binary hidden state, with context affecting hidden inputs. A related mapping records that explicit component/conditional construction without asserting that the entire autoregressive or temporally linked sequence model is one unconditioned RBM. |
Temporal restricted Boltzmann machine
AIMech:TemporalRestrictedBoltzmannMachine |
Architectures | related_match | Each time step uses a conditional RBM with undirected visible–hidden interactions. The entire sequence construction additionally has directed history connections and a separate conditional normalizer per step; a related mapping records the component relationship without equating the whole temporal architecture to one unconditioned symmetric RBM. |
Import Provenance & Upstream Synchronization
Source Repo: AIO GitHub
Release Version:
aio.jsonCommit:
397851c547…License: CC-BY-4.0
Imported At: 2026-10-10T07:34:10Z
SHA256:
8f741d71408c…
Source File:
data/aio/RestrictedBoltzmannMachine.yaml