A Polynomial Time Graph Isomorphism Algorithm via Self-Supervised Gradient Descent
Keywords: Graph Isomorphism, Optimization, Self-Supervised Learning, Graph Matching, Graph Theory
Abstract: Graph isomorphism (GI) is a fundamental problem in graph theory.
Despite recent advancements, determining whether two graphs are isomorphic remains computationally challenging.
This paper introduces the Polynomial Time Graph Isomorphism (PTGI) algorithm, an optimization-based approach leveraging self-supervision techniques to efficiently tackle the graph isomorphism problem.
PTGI aims to escape local optima caused by graph symmetries and provides high accuracy in identifying isomorphic graphs in polynomial time.
Experimental results demonstrate PTGI's effectiveness across various graph types, making it a valuable tool for practical applications.
Primary Area: unsupervised, self-supervised, semi-supervised, and supervised representation learning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 8919
Loading