ALCModule

class mowl.nn.ALCModule(*args: Any, **kwargs: Any)[source]

Bases: Module

Subclass of torch.nn.Module for \(\mathcal{ALC}\) models.

This class provides an interface for neural modules that embed the \(\mathcal{ALC}\) description logic language. In contrast with ELModule, which exposes one loss function per \(\mathcal{EL}\) normal form, \(\mathcal{ALC}\) allows arbitrarily nested class expressions (including negation, disjunction and universal restrictions). Therefore the interface is defined in terms of:

  • forward(), which computes the loss of a (grouped) axiom, and

  • forward_fs(), which recursively computes the fuzzy set membership degree of a class expression over a set of (anonymous) entity embeddings.

The fuzzy logical operators (_logical_and(), _logical_or(), _logical_not(), _logical_exist(), _logical_forall() and _logical_residuum()) are declared here as part of the interface; their concrete semantics (e.g. product t-norm, Łukasiewicz t-norm) are chosen by the subclass. More information can be found at Embedding the EL language.

Methods Summary

forward(axiom, x, e_emb, *args, **kwargs)

Computes the loss of an axiom given an entity-membership context.

forward_fs(cexpr, x, e_emb[, cur_index])

Recursively computes the fuzzy-set membership degree of a class expression over the entities e_emb.

Methods Documentation

forward(axiom, x, e_emb, *args, **kwargs)[source]

Computes the loss of an axiom given an entity-membership context.

Parameters:

  • axiom – The (grouped) axiom or axiom pattern to score. Typically an instance of org.semanticweb.owlapi.model.OWLAxiom.

  • x (torch.Tensor) – Tensor encoding the concrete classes, properties and individuals that fill the axiom pattern.

  • e_emb (torch.Tensor) – Embeddings of the entities (including sampled anonymous entities) over which fuzzy memberships are evaluated.

forward_fs(cexpr, x, e_emb, cur_index=0)[source]

Recursively computes the fuzzy-set membership degree of a class expression over the entities e_emb.

Parameters:

  • cexpr – Class expression of type org.semanticweb.owlapi.model.OWLClassExpression.

  • x (torch.Tensor) – Tensor encoding the named classes/properties of cexpr.

  • e_emb (torch.Tensor) – Entity embeddings.

  • cur_index (int, optional) – Current column offset into x while walking the expression tree. Defaults to 0.