Search papers, labs, and topics across Lattice.
This paper introduces sLTN, an extension of Logic Tensor Networks (LTN) that incorporates structural dimensions to better capture complex relationships in data, such as temporal order and graph connectivity. By treating structural dimensions as first-class elements, sLTN allows for the explicit representation of constraints related to time steps, sequence positions, and relational structures within a neurosymbolic framework. The authors demonstrate that sLTN recovers the original LTN semantics when structural dimensions are absent and provide a PyTorch implementation to facilitate practical applications in reasoning tasks.
sLTN transforms the way we integrate structural information into neurosymbolic reasoning, enabling richer representations of temporal and relational constraints.
Logic Tensor Networks (LTN) provide a neurosymbolic framework in which first-order logic is interpreted through tensor operations, enabling logical constraints to be integrated with differentiable learning. However, the original formulation of LTN is primarily suited to data represented as flat collections of individuals, and does not explicitly capture structural organization such as temporal order, sequential position, or graph connectivity. We introduce sLTN, an extension of LTN that makes structural dimensions first-class elements of the language. Structural dimensions represent named tensor axes associated with domain-specific organization, such as time steps, sequence positions, or graph nodes. They can be quantified explicitly, related through structural relations, and used to express temporal, sequential, and relational constraints directly at the logical level. We formalize the syntax and fuzzy tensor semantics of sLTN and show that, in the absence of structural dimensions, the framework recovers the original LTN semantics as a special case. We further describe a PyTorch implementation based on a declarative signature, formula parsing, and tensorial interpretation. The framework is illustrated on representative temporal and sequential reasoning examples. This paper serves as a companion to the sltn library, available at https://github.com/logictensornetworks/sltn.
