Page 1 of 2

Re: The worst ML paper with physics application I have ever seen

Posted: July 26th, 2020, 12:43 pm
by Cuchulainn
https://arxiv.org/abs/2007.01420

They don't even know about sparse matrix diagonalisation!
ARPACK? 

That article is a train wreck, it's excruciating to read and I don't know what their problem is, let alone their solution.

Anti-pattern: Mr. Miyagi say too many co-authors destroy meal; n persons ==> n(n-1)/2 communication paths. I hope their software is not like that.

Re: The worst ML paper with physics application I have ever seen

Posted: July 26th, 2020, 7:32 pm
by katastrofa
Exact solution by Onsager 1944.

Re: The worst ML paper with physics application I have ever seen

Posted: July 27th, 2020, 11:20 am
by Cuchulainn

Re: The worst ML paper with physics application I have ever seen

Posted: August 7th, 2020, 11:50 pm
by katastrofa
Have you heard about deep neural networks for data augmentation? I've made one for you!
F0F58084-6FF7-4F70-9A22-48A2E0B51F77.jpeg

Re: The worst ML paper with physics application I have ever seen

Posted: August 8th, 2020, 8:08 pm
by bearish
Is that von Neumann’s elephant?

Re: The worst ML paper with physics application I have ever seen

Posted: August 8th, 2020, 8:48 pm
by Cuchulainn
Image

Re: The worst ML paper with physics application I have ever seen

Posted: August 8th, 2020, 8:58 pm
by Cuchulainn
We need more plumbers (the day after I presented 1st edition of my C++ book)

Image

Re: The worst ML paper with physics application I have ever seen

Posted: August 9th, 2020, 1:35 am
by katastrofa
Nice hat!

I've never been good at drawing Neumann's elephants - instead of arts classes, they taught me to shot and throw grenades instead. Not so good at plumbing either :-(
I think what is crucial about the elephant is whether he or she is a pattern or just a fluke - or, to put it differently, am I looking at the elephant from the ant's or the grand scheme of things' perspective. All the deep stuff and we're back to square 1.

Re: The worst ML paper with physics application I have ever seen

Posted: August 9th, 2020, 9:42 am
by Cuchulainn
You forgot the made-to-measure Scots tweed suit (from West End Tailors, Amsterdam).

Re: The worst ML paper with physics application I have ever seen

Posted: August 9th, 2020, 8:11 pm
by katastrofa
Are you really Irish? You have an American accent and... do you pluck your eyebrows??

Re: The worst ML paper with physics application I have ever seen

Posted: August 10th, 2020, 9:50 am
by Cuchulainn
Our family came from Scotland but we changed the name to make it sound more Irish.
And my dad was American Irish.
Junior is my barber.He's learning his trade. And a fully qualified cabinet maker and composer.

Re: The worst ML paper with physics application I have ever seen

Posted: August 10th, 2020, 1:17 pm
by katastrofa
Tinker, tailor, soldier, spy.

Re: The worst ML paper with physics application I have ever seen

Posted: August 10th, 2020, 3:54 pm
by Cuchulainn
One for the money
Two for the show
Three to get ready
And four to go.

Re: The worst ML paper with physics application I have ever seen

Posted: August 17th, 2020, 2:26 pm
by Cuchulainn
Here's a screamer

https://arxiv.org/pdf/1611.04135.pdf


In November 2016 we submitted to arXiv our paper “Automated Inference on Criminality Using Face Images”. It generated a great deal of discussions in the Internet and some media outlets. Our work is only intended for pure academic discussions; how it has become a media consumption is a total surprise to us. Although in agreement with our critics on the need and importance of policing AI research for the general good of the society, we are deeply baffled by the ways some of them mispresented our work, in particular the motive and objective of our research.

Re: The worst ML paper with physics application I have ever seen

Posted: August 26th, 2020, 10:54 am
by Cuchulainn
And this?

Image