$Graph Embedding via Topology and Functional Analysis$

Phani raj Chinnalingu

Keywords: Graph embedding, Theory, Topology, Functional analysis

Abstract: Graphs have been ubiquitous in Machine Learning due to their versatile nature in modelling real world situations .Graph embedding is an important precursor to using graphs in Machine Learning , and much of performance of algorithms developed later depends heavily on this. However very little theoretical work exists in this area , resulting in the proliferation of several benchmarks without any mathematical validation , which is detrimental .In this paper we present an analysis of deterministic graph embedding in general , using tools from Functional Analysis and Topology . We prove several important results pertaining to graph embedding which may have practical importance .One limitation of our work in it's present form is it's applicable to deterministic embedding approaches only, although we strongly hope to extend it to random graph embedding methods as well in future.We sincerely hope that this work will be beneficial to researchers working in field of graph embedding.

One-sentence Summary: This paper some results corresponding to the theory of deterministic graph embedding are given and also at places their practical significance is highlighted

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

Reviewed Version (pdf): https://openreview.net/references/pdf?id=qZiPEe3MBr