![](https://secure.gravatar.com/avatar/91dc879a1f4f8a107be62f4e108002de.jpg?s=120&d=mm&r=g)
Hello, I find all the ideas posted in reply to my message very interesting, thank you very much to all who have answered to my question. Especially, I would like to know more about Kilian's and Robin's suggestions. In particular, I find difficult to understand and translate the ideas posted by them into my background. Of course, this is not Kilian's or Robin's fault, but my complete fault due to lack of knowledge. To Robin: - Is there a paper covering your package, but explained in layman's terms, not requiring previous knowledge on the subject? Or maybe a simple but fully-worked example (ideally closely related to the Ising model) that can be used in your package to see how everything works. To Kilian: - Do you have a computer package that covers that computations in your paper? Or do you have the Ising code available to distribution? I would be very interested to know more about the Ising implementation of your paper. Kind regards Jordi
CC: jordi_molins@hotmail.com From: koepsell@gmail.com To: scipy-user@scipy.org Subject: Re: [SciPy-User] [SciPy-user] Maximum entropy distribution for Ising model - setup? Date: Mon, 11 Jan 2010 18:02:53 -0800
Jordi,
On Jan 7, 2010, at 1:09 AM, Jordi Molins Coronado wrote:
Hello, I am new to this forum. I am looking for a numerical solution to the inverse problem of an Ising model (or a model not- unlike the Ising model, see below). I have seen an old discussion, but very interesting, about this subject on this forum (http://mail.scipy.org/pipermail/scipy-user/2006-October/009703.html ).
You might want to check out a recent method developed in our group, called "Minimum Probability Flow Learning" that allows very fast parameter estimation of basically any distribution -- including the Ising model. A 100 unit ising model can be fitted within about 1 minute (see Fig. 3). The paper is here: http://arxiv.org/abs/0906.4779
Kilian
-- Kilian Koepsell, PhD Redwood Center for Theoretical Neuroscience Helen Wills Neuroscience Institute, UC Berkeley 156 Stanley Hall, MC# 3220 , Berkeley, CA 94720