English

A non-negative expansion for small Jensen-Shannon Divergences

Machine Learning 2008-10-29 v1

Abstract

In this report, we derive a non-negative series expansion for the Jensen-Shannon divergence (JSD) between two probability distributions. This series expansion is shown to be useful for numerical calculations of the JSD, when the probability distributions are nearly equal, and for which, consequently, small numerical errors dominate evaluation.

Keywords

Cite

@article{arxiv.0810.5117,
  title  = {A non-negative expansion for small Jensen-Shannon Divergences},
  author = {Anil Raj and Chris H. Wiggins},
  journal= {arXiv preprint arXiv:0810.5117},
  year   = {2008}
}

Comments

4 page technical report, 2 figures