Serving the Quantitative Finance Community

 
User avatar
Cuchulainn
Posts: 64390
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: DL and PDEs

September 10th, 2019, 8:29 am

Looks like someone's 20% project.

This project work seems to be part of Tensorflow (pip --upgrade Tensorflow). Strange brew?
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64390
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: DL and PDEs

October 25th, 2019, 3:20 pm

Another one..

https://arxiv.org/abs/1910.10262

Cute ODE (10)
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
JohnLeM
Posts: 515
Joined: September 16th, 2008, 7:15 pm

Re: DL and PDEs

October 25th, 2019, 4:13 pm

Another one..

https://arxiv.org/abs/1910.10262

Cute ODE (10)
Another one ... bite the dust ?
mmm...I am really feeling some kind of embarrassments when I look to all these DL papers that will revolution PDE methods.
 
User avatar
Cuchulainn
Posts: 64390
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: DL and PDEs

October 25th, 2019, 4:35 pm

That's just ruined my weekend

www.youtube.com/watch?v=fnZDI1slxMI
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64390
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: DL and PDEs

April 11th, 2020, 1:52 pm

How to write a good technical article (Steve McConnell)

https://stevemcconnell.com/articles/how ... l-article/
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64390
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: DL and PDEs

August 14th, 2020, 4:15 pm

This article is intriguing; 

https://arxiv.org/abs/1804.04272

This slide show addresses a number of shortcoming with traditional deep learning models.

https://gateway.newton.ac.uk/sites/defa ... amer_0.pdf

I'll post some remarks in the coming days.
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64390
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: DL and PDEs

September 9th, 2020, 9:25 am

Can ML produce models directly from data?

Given enough time, a monkey could write the complete works of Shakespeare and in more serious vein if we had enough training data could we recover the incompressible Navier-Stokes PDE, for example?

You have models, parameters of model and you have data.
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl
 
User avatar
Cuchulainn
Posts: 64390
Joined: July 16th, 2004, 7:38 am
Location: Drosophila melanogaster
Contact:

Re: DL and PDEs

April 9th, 2021, 10:05 am

Did you guys discuss this here already?
https://arxiv.org/pdf/1706.04702.pdf
Update: problem solved.

A proof that artificial neural networks overcome the curse of dimensionality in the numerical approximation of Black-Scholes partial differential equations

https://arxiv.org/abs/1809.02362

And if you believe that, you'll believe anything.
"Compatibility means deliberately repeating other people's mistakes."
David Wheeler

http://www.datasimfinancial.com
http://www.datasim.nl