Skip to main content
6 events
when toggle format what by license comment
Oct 10, 2019 at 20:00 comment added meh It's worth noting that KL is typically called divergence and not metric because it is not symmetric !
S Oct 8, 2019 at 11:53 history suggested Ivo Merchiers CC BY-SA 4.0
Fix broken link to wikipedia
Oct 8, 2019 at 11:17 review Suggested edits
S Oct 8, 2019 at 11:53
May 3, 2010 at 21:41 comment added marc In evolutionary game theory the Fisher information metric is called the Shahshahani metric. See "Evolutionary Game Theory" by Hofbauer and Sigmund. The information divergence can be used to give a Lyapunov function for the replicator equation. I believe this goes back to Akin; others use an exponentiated form (e.g. Hofbauer and Sigmund).
Apr 30, 2010 at 22:23 comment added Suresh Venkat I'd be interested in references to the work in evolutionary game theory. As an aside, the KL divergence is used extensively in machine learning, as well as in computer vision and (to a degree) in information retrieval.
Apr 30, 2010 at 19:56 history answered marc CC BY-SA 2.5