We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.DM

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Discrete Mathematics

Title: Semantic Embeddings in Semilattices

Abstract: To represent anything from mathematical concepts to real-world objects, we have to resort to an encoding. Encodings, such as written language, usually assume a decoder that understands a rich shared code. A semantic embedding is a form of encoding that assumes a decoder with no knowledge, or little knowledge, beyond the basic rules of a mathematical formalism such as an algebra. Here we give a formal definition of a semantic embedding in a semilattice which can be used to resolve machine learning and classic computer science problems. Specifically, a semantic embedding of a problem is here an encoding of the problem as sentences in an algebraic theory that extends the theory of semilattices. We use the recently introduced formalism of finite atomized semilattices to study the properties of the embeddings and their finite models. For a problem embedded in a semilattice, we show that every solution has a model atomized by an irreducible subset of the non-redundant atoms of the freest model of the embedding. We give examples of semantic embeddings that can be used to find solutions for the N-Queen's completion, the Sudoku, and the Hamiltonian Path problems.
Comments: 58 pages
Subjects: Discrete Mathematics (cs.DM); Logic in Computer Science (cs.LO); Commutative Algebra (math.AC); Logic (math.LO); Rings and Algebras (math.RA)
Cite as: arXiv:2205.12618 [cs.DM]
  (or arXiv:2205.12618v1 [cs.DM] for this version)

Submission history

From: Fernando Martin-Maroto [view email]
[v1] Wed, 25 May 2022 09:55:17 GMT (44kb)

Link back to: arXiv, form interface, contact.