Hi Tom and all, This looks interesting. I think there may be some ways to link the variational approach to this. I have looked at the variational approach and done a few experiments. There are a number of issues to look into but I think it is promising so far. Ciira On Tue, Sep 6, 2022 at 6:20 PM Dietterich, Thomas <tgd@oregonstate.edu> wrote:
I spent the weekend running experiments on the latent precipitation model. Instead of fitting the model with Bernoulli noise, I fit it with Gaussian noise (which is tractable). It works pretty well on the Oklahoma data. We can discuss this on Wednesday.
--Tom
Thomas G. Dietterich, Distinguished Professor Voice: 541-737-5559
School of Electrical Engineering FAX: 541-737-1300
and Computer Science URL: eecs.oregonstate.edu/~tgd
US Mail: 1148 Kelley Engineering Center
Office: 2067 Kelley Engineering Center
Oregon State Univ., Corvallis, OR 97331-5501
-- Rainqc-jobman mailing list Rainqc-jobman@ENGR.ORST.EDU https://it.engineering.oregonstate.edu/mailman/listinfo/rainqc-jobman
-- Ciira Maina, Ph.D. Snr. Lecturer, Department of Electrical and Electronic Engineering, Director, Centre for Data Science and Artificial Intelligence (DSAIL), Dedan Kimathi University of Technology. Nyeri-Kenya Webpage: http://www.ciirawamaina.com/ Blog: http://ciirawamaina.com/blog/blog.html