Posted on 05/18/2019 4:29:11 PM PDT by BenLurkin
You’ve got this wrong
There was a breakthrough several years ago in machine learning.
Google “deep learning” for a starting point.
Or at least the correct output that it thinks is correct, is obtained. That means it may still be wrong.
I believe that the princiiples of Philo Farnsworths fusor could be tweeked to produce net positive energy, but no one has been clever enough to do this...yet.
Its been 20 years away for the last 60 years.
Yup. It’s telling how wide of the mark 2001 A Space Odyssey was. We’re not where we thought we would be.
“Artificial Intelligence Accelerates Development of Limitless Fusion Energy”
yeah, right. they can’t even get AI to translate a webpage accurately from one language to another, but pretty soon an AI will build a fusion reactor that no human has been able to build ...
Now even machine AI looks for a predetermined answer
A “fussion reaction”?
Is that a hybrid of fusion and fission? Must be some sort of breakthrough that I haven’t heard of before.
But look at how much easier AI has made our lives when calling for help.
IVR: “Listen carefully as our menu has changed. Due to high call volume, we will answer your all in four days, three hours, and twenty seven minutes.”
IVR: “Please say the name of the person or department you are trying to reach.”
Me: “Doctor Bill Fried.”
IVR: “I think you said you want the Billing Department. Is that right?”
Me: “No. Doctor Bill Fried.”
IVR: “I think you said you had fried eggs for breakfast. Is that right?”
Me. “NO! I did not have fried eggs for breakfast. I want Fried! Doctor Fried!!”
IVR: “Do you need to be freed from captivity? If this is an emergency, hang up and call 911.”
Me: “NO NO NO You effing robot. I want the doctor named Bill Fried.”
IVR (nauseating chirpy happy voice): “OK, I will connect you to the billing department. They can free you.”
Note: this topic is from . Thanks BenLurkin.
Ding, ding, ding - we have a thread winnah!
when a node fails to achieve this task: the weights automatically adjust themselves for fresh data until the correct output is obtained.
......
if this performs as expected —it will greatly accelerate the time when fusion reactors become available.
They have been getting closer to ignition in the last four years. But its still a physics problem. Meaning they havn’t figured out the physics yet. Meaning that its still 20 years off.
However, if the whole process of experimentation could be accelerated byo machine learning as is done with materials research and dna research—then the time to net positive energy output will be shrunk.
Of course, we don’t know if the machine learning actually works or whether it solves for the right things.
So we’ll see.
I'm sure it'll prove to be a big help in the continued development of virtual fusion reactors. :^)
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.