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

Download:

Current browse context:

cs.LG

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 > Machine Learning

Title: Learning Combinatorial Optimization on Graphs: A Survey with Applications to Networking

Abstract: Existing approaches to solving combinatorial optimization problems on graphs suffer from the need to engineer each problem algorithmically, with practical problems recurring in many instances. The practical side of theoretical computer science, such as computational complexity then needs to be addressed. Relevant developments in machine learning research on graphs is surveyed, for this purpose. We organize and compare the structures involved with learning to solve combinatorial optimization problems, with a special eye on the telecommunications domain and its continuous development of live and research networks.
Comments: 27 pages, 0 figures, submitted to IEEE Access journal
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
Cite as: arXiv:2005.11081 [cs.LG]
  (or arXiv:2005.11081v1 [cs.LG] for this version)

Submission history

From: Natalia Vesselinova [view email]
[v1] Fri, 22 May 2020 09:45:36 GMT (3600kb,D)

Link back to: arXiv, form interface, contact.