Posted on 05/18/2019 4:29:11 PM PDT by BenLurkin
…[A] team of scientists working with a Harvard graduate student is for the first time applying deep learning — a powerful new version of the machine learning form of AI — to forecast sudden disruptions that can halt fusion reactions and damage the doughnut-shaped tokamaks that house the reactions.
Unlike traditional software, which carries out prescribed instructions, deep learning learns from its mistakes. Accomplishing this seeming magic are neural networks, layers of interconnected nodes — mathematical algorithms — that are “parameterized,” or weighted by the program to shape the desired output. For any given input the nodes seek to produce a specified output, such as correct identification of a face or accurate forecasts of a disruption. Training kicks in when a node fails to achieve this task: the weights automatically adjust themselves for fresh data until the correct output is obtained.
A key feature of deep learning is its ability to capture high-dimensional rather than one-dimensional data. For example, while non-deep learning software might consider the temperature of a plasma at a single point in time, the FRNN considers profiles of the temperature developing in time and space.
The vast databases have enabled reliable predictions of disruptions on tokamaks other than those on which the system was trained — in this case from the smaller DIII-D to the larger JET. The achievement bodes well for the prediction of disruptions on ITER, a far larger and more powerful tokamak that will have to apply capabilities learned on today’s fusion facilities.
(Excerpt) Read more at scitechdaily.com ...
Call me when I can purchase my carbon nano-tube flying car.
Sounds promising.
I hope it works and it’s not just some anomaly.
“Any sufficiently advanced technology is indistinguishable from magic.”
—Arthur C Clarke
Sounds a lot like Democrat vote counting.
It’s just around the corner.
Fusion —> Free Helium for everyone....
Probably in 10 years...
*HAL: I'm sorry Dave, I can't do that.
I needs me some of that deep learning, Uncle Jed.
This is the cultural-marxists worst nightmare, free citizens becoming autonomous, and independent of state welfare.
After all, you can’t be a prince if you don’t have serfs.
Uh, it is so basic to realize distributed or parallel processing is needed that this tells me these people are pretty stupid.
I took a graduate distributed and parallel programming class as an undergraduate and did greatly trimmed down versions of this in 2001.
It has been for decades.
Just around the corner...
Still...
That was the idea, just like nuclear fusion reactors have been around the corner for sixty years.
I’ll believe it’s here when I can buy a home version at Harbor Freight.
It will work for a couple of days or several years.
Hard to tell with their stuff. Some tools I’ve bought from there have done amazingly well. Others went in the trash within a day.
My frat held the first Tokamak party ever.
HYPE
PLEASE GIVE US MORE talk about What has Already Actually BEEN Accomplished, instead of what they SAY they are GOING TO do.
Claims were made over 60 years ago about what Artificial Intelligence was going to ‘soon do’ and little of the claims has ever come true.
The problem with AI is that any significantly complex logic/knowledge has to come from a human, which is horribly slow and flawed process, and the midway results have to be constantly checked (the machine told its right or wrong) - again by a horribly slow human expert.
Most of what is called AI is merely search filtering - a useful tool, but that is all it is - a tool - still being used/controlled by a human mind.
Does China control the dilithium crystal market?
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.