Artificial Intelligence Takes On Earthquake Prediction

Binary_Bark

Binary_Bark

Forging
Joined
Feb 24, 2016
Messages
42,032
Reaction score
24,080
Location
Midgard
In May of last year, after a 13-month slumber, the ground beneath Washington’s Puget Sound rumbled to life. The quake began more than 20 miles below the Olympic mountains and, over the course of a few weeks, drifted northwest, reaching Canada’s Vancouver Island. It then briefly reversed course, migrating back across the U.S. border before going silent again. All told, the monthlong earthquake likely released enough energy to register as a magnitude 6. By the time it was done, the southern tip of Vancouver Island had been thrust a centimetre or so closer to the Pacific Ocean.

Because the quake was so spread out in time and space, however, it’s likely that no one felt it. These kinds of phantom earthquakes, which occur deeper underground than conventional, fast earthquakes, are known as “slow slips.” They occur roughly once a year in the Pacific Northwest, along a stretch of the fault where the Juan de Fuca plate is slowly wedging itself beneath the North American plate. More than a dozen slow slips have been detected by the region’s sprawling network of seismic stations since 2003. And for the past year and a half, these events have been the focus of a new effort at earthquake prediction by the geophysicist Paul Johnson.
Click to expand...

More At: https://www.quantamagazine.org/artificial-intelligence-takes-on-earthquake-prediction-20190919/
 
Random Tree Forests to predict earthquakes :D . Pretty powerful algorithms.
 
I am (currently) in IT, and from time to time do some programming.
I have no idea what the hell AI is.
How is AI different from a/any algorithm?
 
Ponderer said:
Care to explain how AI is different from a/any algorithm?
Click to expand...

AI (specifically machine learning algorithms), learn how to make the best predictions using parameters derived from data that it is trained on.
 
cguy said:
AI (specifically machine learning algorithms), learn how to make the best predictions using parameters derived from data that it is trained on.
Click to expand...
What is a machine learning algorithm?
How be it different from a regular algorithm?
 
Ponderer said:
What is a machine learning algorithm?
How be it different from a regular algorithm?
Click to expand...

Machine learning is the most popular type of AI algorithm today. Older ones try to do things such as trying to mimic human behavior by having a programmer exhaustively try replicate human like decision logic.

Regular algorithms don’t learn their parameters from training data.
 
cguy said:
Regular algorithms don’t learn their parameters from training data.
Click to expand...
No matter how you try and spin it, its nothing other than an algorithm.
Algorithms do not and cannot write themselves.
 
Ponderer said:
Care to explain how AI is different from a/any algorithm?
Click to expand...
AI is different to machine learning as ML uses fixed instruction sets to achieve an outcome where AI can recode all its instruction sets for a completely different outcome. AI needs vastly more processing power. You guys should know this.
 
Magnum said:
AI is different to machine learning as ML uses fixed instruction sets to achieve an outcome where AI can recode all its instruction sets for a completely different outcome. AI needs vastly more processing power. You guys should know this.
Click to expand...
Are you contending that an algorithm can write itself?
 
Ponderer said:
No matter how you try and spin it, its nothing other than an algorithm.
Algorithms do not and cannot write themselves.
Click to expand...

Nobody said it’s anything but an algorithm. It’s a particular subclass of algorithm, that learns its parameters.
 
Magnum said:
Yes. it only gets the basic syntax as fixed information. after that its free for all.
Click to expand...
You a programmer?
Those who are programmers would know you are talking out of your a$$.
 
Magnum said:
AI is different to machine learning as ML uses fixed instruction sets to achieve an outcome where AI can recode all its instruction sets for a completely different outcome. AI needs vastly more processing power. You guys should know this.
Click to expand...
Machine learning is a subset of AI, not sure what you mean by the above sentence.

Ponderer said:
No matter how you try and spin it, its nothing other than an algorithm.
Algorithms do not and cannot write themselves.
Click to expand...
An algorithm that constantly adjust another algorithm which maps the relationship between input and output, which can be very complex. In essense everything on a computer is an algorithm or multiple algorithms, it's just that these are for a more specific type of task.
 
cguy said:
Nobody said it’s anything but an algorithm. It’s a particular subclass of algorithm, that learns its parameters.
Click to expand...
Its still an algorithm mate.
Algorithms do not and cannot write themselves.
 
no Deep blue was insanely powerful. But it could only do 1 thing. play chess. It could calculate thousands of moves per second, basically every possible move there is and the best counter to it. But It could only play chess. Apply that to a algorithm that changes into what it needs to be. and you have a true AI.
Therefor AI is an algorithm that rewrites itself to perform a task faster, better and with no previous instruction set.

The problem is with the 3 machine fundamental rules. there are multiple ways to surcomvent them. if you software code them, The AI could just duplicate without them. if you hard code them into hardware the AI could create a soft copy and build a new Hardware platform without the limitation.
 
Ponderer said:
Its still an algorithm mate.
Algorithms do not and cannot write themselves.
Click to expand...

You are correct, that an algorithm is indeed still an algorithm.

Learning and “writing themselves” are two different things. Machine learning algorithms do in fact write algorithms, but they do not write themselves - what would be the point, given that they are already written?
 
Magnum said:
no Deep blue was insanely powerful. But it could only do 1 thing. play chess. It could calculate thousands of moves per second, basically every possible move there is and the best counter to it. But It could only play chess. Apply that to a algorithm that changes into what it needs to be. and you have a true AI.
Therefor AI is an algorithm that rewrites itself to perform a task faster, better and with no previous instruction set.
Click to expand...
Yup - talking out of your a$$.
 
You must log in or register to reply here.
Back
Top
Sign up to the MyBroadband newsletter
X