Deep neural networks are revolutionizing astronomy and many other things.

 Deep neural networks are revolutionizing astronomy and many other things.

Deep learning means that the person would know how to act in certain situations. "Learning" means. The person would know how to in any situation that connects with a certain case.

But deep learning means that the actor also knows why something is done in a certain case?  And also, the actor can predict some situations. In the computer world, the prediction means that the solution like some movement series will upload to RAM (Read Access Memory for) immediate use. 

Deep neural networks in the service of astronomers are opening the road for a new type of artificial intelligence. The idea of the deep neural network is searching the exoplanets is that the system is following a certain well-known exoplanet using multiple different types of telescopes.

And then, that system would make the database about the observations. Then the network is trying to look for the phenomenon that is matching with recorded data to other stars. The ability to collect the data also by using the smaller telescope simulates the targets. Those are farther in the universe than the well-known exoplanet. 

So the small telescope can simulate the situation where some exoplanet locates very far away from the Earth. Then this data can use as the data matrix for the larger size telescopes. And by using the larger telescope. The AI can see if the data that is collected from other and more distant stars match the data. That collected from closer stars. 

The model where the system makes the matrix about the case and then transfers that to other cases would be revolutionary in astronomy. The fact is that this kind of solution can benefit also in many other situations like AI-controlled cars or similar things. If some solutions would be a useful thing. The computer is loading the data from the environment and then the computer would load this matrix in its memory. 

And then in the future, when the automatic would drive to a similar environment the system would load the case matrix to the RAM (Read Access Memory) that the system would use that data matrix or solution immediately. That means the image of the environment or the position on the map would act as a trigger. That uploads the solution for immediate use. This is the computer-world version of the learning. 

So the learning system can predict the situation. When the system records the place, And if there is a match with some case that system preloads the solution or the movement series to its RAM. And that thing means the machine learns to predict things. Machine learning is the thing that is one of the most interesting things in computing. 

In a chess program, machine learning means that the system can record the games of the masters. And then it would make the database for each movement. Then the system must just make the counter-action against the opponent simply by reconnecting the databases. That thing makes the machine can learn things. Like what buttons are the most effective against a certain player. The AI can calculate how often some chess player is moving a certain button. 

And conclude what kind of role the button is the button for that certain player. The chess game is a useful thing to test how to connect databases. 

But the same way the ability to make spontaneous connections can test by using astronomical objects as the base. And that kind of system can help to benefit telescopes, chess, and the AI for finding the new and more independently operating robots and AI a reality.

https://scitechdaily.com/a-whopping-301-newly-confirmed-exoplanets-discovered-with-new-deep-neural-network-exominer/ 

https://interestandinnovation.blogspot.com/