Artificial Intelligence Developed to Predict the Future: What Does It Mean?

Scientists at Ohio State University, USA, developed an algorithm that works even on Windows 10; By looking at similar problems and solutions in the past, he was able to “predict the future of any process, even for 1 second”.

The idea of ​​going to the future is impossible with the technology we have now. Because scientific studies are not related to the behavior of “seeing the future”, which is mostly identified with fortune telling. When you read the titleWhat’s wrong with predicting 1 second ahead?‘ but believe me, it’s not like that. With this discovery made in the world of science, the door of a very important era is opened: Predicting the future.

Researchers from Ohio State University are using a self-developed artificial intelligence called ‘next generation reservoir computing’. spatiotemporal (spatiotemporal) of chaotic systems They found a new way to predict their behavior.

An algorithm that can work even on a Windows 10 laptop looks at complex problems in the past and tries to solve them in the future:

The researchers tested their new algorithm on a complex problem that had been studied many times in the past. This new algorithm, which is supposed to predict the behavior of an atmospheric weather model, is compared to traditional machine learning algorithms that can solve the same problem. It used 400 to 1,250 times less training data.

Moreover, the algorithm achieved this at a much less cost. This algorithm, which does not require supercomputers like traditional machine learning algorithms, Windows 10iIt can be run even on a laptop with an operating system and compared to its competitors 240,000 times faster somehow it can predict the next 1 second.

Wendson De Sa Barbosa, one of the developers of the algorithm, says: “This is very exciting because we believe this is a significant advance in machine learning in terms of data processing efficiency and prediction accuracy.”

Okay, but what will this do for us?

The key thing here is not that the algorithm predicts the next 1 second, but Ability to solve spatio-temporal chaotic systems very quickly. De Sa Barbosa says: “Modern machine learning algorithms are well suited for predicting dynamical systems by learning their basic physical rules using historical data. Once you have enough data and computational power, you can make predictions about any complex system in the real world with machine learning models.”

The biggest example of this is human heart cells. Heart cells are chaotic, switching from a normal heartbeat to an abnormally high frequency spatio-spatial schemes they exhibit. In other words, as a result of this research, we can prevent a heart disease in the future together with artificial intelligence. can be controlled before they show any symptoms means:

“If the equations that accurately describe how these unique processes will evolve for a system are known, its behavior can be reproduced and predicted.”

RELATED NEWS

Artificial Intelligence Is Now Into Politics: So This Is How They Will Rule The World

Here the heart cells are just one example, too many elements with spatiotemporal chaotic problems and schemes exist in the real world. What are you thinking? Please do not forget to share your thoughts with us in the comments.

Source :
https://news.osu.edu/machine-learning-helps-scientists-peer-a-second-into-the-future/


source site-33