Some Remarks on Diffusion Distances

As a diffusion distance, we propose to use a metric (closely related to cosine similarity) which is defined as the 𝐿2 distance between two 𝐿2-normalized vectors. We provide a mathematical explanation as to why the normalization makes diffusion distances more meaningful. Our proposal is in contrast t...

Full description

Saved in:
Bibliographic Details
Main Authors: Maxim J. Goldberg, Seonja Kim
Format: Article
Language:English
Published: Wiley 2010-01-01
Series:Journal of Applied Mathematics
Online Access:http://dx.doi.org/10.1155/2010/464815
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1832554369117061120
author Maxim J. Goldberg
Seonja Kim
author_facet Maxim J. Goldberg
Seonja Kim
author_sort Maxim J. Goldberg
collection DOAJ
description As a diffusion distance, we propose to use a metric (closely related to cosine similarity) which is defined as the 𝐿2 distance between two 𝐿2-normalized vectors. We provide a mathematical explanation as to why the normalization makes diffusion distances more meaningful. Our proposal is in contrast to that made some years ago by R. Coifman which finds the 𝐿2 distance between certain 𝐿1 unit vectors. In the second part of the paper, we give two proofs that an extension of mean first passage time to mean first passage cost satisfies the triangle inequality; we do not assume that the underlying Markov matrix is diagonalizable. We conclude by exhibiting an interesting connection between the (normalized) mean first passage time and the discretized solution of a certain Dirichlet-Poisson problem and verify our result numerically for the simple case of the unit circle.
format Article
id doaj-art-7f7a963f0c1e4ac6bb55c0be368fffc9
institution Kabale University
issn 1110-757X
1687-0042
language English
publishDate 2010-01-01
publisher Wiley
record_format Article
series Journal of Applied Mathematics
spelling doaj-art-7f7a963f0c1e4ac6bb55c0be368fffc92025-02-03T05:51:41ZengWileyJournal of Applied Mathematics1110-757X1687-00422010-01-01201010.1155/2010/464815464815Some Remarks on Diffusion DistancesMaxim J. Goldberg0Seonja Kim1Theoretical and Applied Science, Ramapo College of NJ, 505 Ramapo Valley Road, Mahwah, NJ 07430, USAMathematics Department, SUNY Rockland Community College, 145 College Road, Suffern, NY 10901, USAAs a diffusion distance, we propose to use a metric (closely related to cosine similarity) which is defined as the 𝐿2 distance between two 𝐿2-normalized vectors. We provide a mathematical explanation as to why the normalization makes diffusion distances more meaningful. Our proposal is in contrast to that made some years ago by R. Coifman which finds the 𝐿2 distance between certain 𝐿1 unit vectors. In the second part of the paper, we give two proofs that an extension of mean first passage time to mean first passage cost satisfies the triangle inequality; we do not assume that the underlying Markov matrix is diagonalizable. We conclude by exhibiting an interesting connection between the (normalized) mean first passage time and the discretized solution of a certain Dirichlet-Poisson problem and verify our result numerically for the simple case of the unit circle.http://dx.doi.org/10.1155/2010/464815
spellingShingle Maxim J. Goldberg
Seonja Kim
Some Remarks on Diffusion Distances
Journal of Applied Mathematics
title Some Remarks on Diffusion Distances
title_full Some Remarks on Diffusion Distances
title_fullStr Some Remarks on Diffusion Distances
title_full_unstemmed Some Remarks on Diffusion Distances
title_short Some Remarks on Diffusion Distances
title_sort some remarks on diffusion distances
url http://dx.doi.org/10.1155/2010/464815
work_keys_str_mv AT maximjgoldberg someremarksondiffusiondistances
AT seonjakim someremarksondiffusiondistances