Experiments with noisy kriging for the latent precipitation model
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
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
Hi Ciira, I have made a few observations about the proposed GP with Bernoulli noise model. Please take a look and we can discuss at the end of today's rainqc meeting if you are attending. Cheers, Tema On Wed, Sep 7, 2022 at 5:05 AM Ciira Maina <cwamaina.dekut@gmail.com> wrote:
[This email originated from outside of OSU. Use caution with links and attachments.] 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 -- Rainqc-jobman mailing list Rainqc-jobman@ENGR.ORST.EDU https://it.engineering.oregonstate.edu/mailman/listinfo/rainqc-jobman
Thanks Tema, These are useful comments. I'll try to handle them when we have our meeting next week. I can't attend today. On Wed, Sep 14, 2022 at 8:50 AM Tematelewo, Irene <temateli@oregonstate.edu> wrote:
Hi Ciira,
I have made a few observations about the proposed GP with Bernoulli noise model.
Please take a look and we can discuss at the end of today's rainqc meeting if you are attending.
Cheers, Tema
On Wed, Sep 7, 2022 at 5:05 AM Ciira Maina <cwamaina.dekut@gmail.com> wrote:
[This email originated from outside of OSU. Use caution with links and attachments.] 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 -- Rainqc-jobman mailing list Rainqc-jobman@ENGR.ORST.EDU https://it.engineering.oregonstate.edu/mailman/listinfo/rainqc-jobman
-- 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
participants (3)
-
Ciira Maina -
Dietterich, Thomas -
Tematelewo, Irene