Talk:Determinantal point process

This article needs some cleaning up. I don't have the time at the moment nor do I know much about said point processes. That said, some suggestions:

• A softer or easier introductory

• more details eg write that "det" denotes the determinant, include a definition or link to kernel (as this is quite a generic term without context).

• Determinantal point processes (DPP) generalize a number of point processes such Poisson, Strauss and others.

• Perhaps get some ideas from Terence Tao's blog entry

• make it clear that DPPs can have a variable degree of repulsion between points, which explains their use as models of Fermions (corresponding to the Paul-Exlcusion Principle)

• name current or proposed applications (in addition to their original uses as particle models in physics) eg they have been proposed to be used in machine learning by Kulesza and Taskar, they are being used to model wireless communications, especially cellular networks where there is "repulsion" between transmitters.

• regarding the condition of existence, isn't the first condition (symmetry) of the joint intensity or correlation function automatically met? the correlation function is the derivative of a factorial measure, which is symmetric (?), hence its derivative/density is symmetric too?

Improbable keeler (talk) 14:52, 6 November 2013 (UTC)

Reference "Lyons, R. with Peres, Y., Probability on Trees and Networks. Cambridge University Press" should be removed as it is rather about other topic. Even word "determinantal" is used only 3 times, at page 135 of current version. I could not find significant intersection with the current page topic, though the authors have elsewhere (in their papers) discussed it.Zoran.skoda (talk) 17:53, 3 May 2014 (UTC)

Add A Fact: "DPPs in machine learning"

I found a fact that might belong in this article. See the quote below

Determinantal point processes (DPPs) offer a promising and complementary approach. Arising in quantum physics and random matrix theory, DPPs are elegant probabilistic models of global, negative correlations, and offer efficient algorithms for sampling, marginalization, conditioning, and other inference tasks

The fact comes from the following source:

https://ar5iv.labs.arxiv.org/html/1207.6083

Here is a wikitext snippet to use as a reference:

 {{Cite web |title=Determinantal point processes for machine learning |url=https://ar5iv.labs.arxiv.org/html/1207.6083 |website=ar5iv |access-date=2024-10-13 |language=en |quote=Determinantal point processes (DPPs) offer a promising and complementary approach. Arising in quantum physics and random matrix theory, DPPs are elegant probabilistic models of global, negative correlations, and offer efficient algorithms for sampling, marginalization, conditioning, and other inference tasks}} 

This post was generated using the Add A Fact browser extension.

pony in a strange land (talk) 01:12, 13 October 2024 (UTC)