The human brain contains more than 100 billion neurons, and it has an enormous and unlimited capacity to collect and store information.
Scientists have found that the volume of information that the human mind can collect and store, within only one year, reaches 300 exabytes (Exabyte). To demonstrate this capacity, memorizing all words in all languages used worldwide occupies only 5 exabytes, and this gives an idea of the huge capabilities of the human mind.
However, this human brain is a selective organism, it chooses forgetfulness often, and puts large amounts of information that it categorizes as insignificant with discarded files in convolutions of the mind, there is a lot of information that is stuffed into our mind from birth to death, both in school, university, work, or through our daily life movement.
We encounter large numbers of people in our lives, but we often fail to remember them, for example, we may meet someone, and he comes to shake our hand very warmly which suggests that he knows us completely, and yet we hardly remember his name or where we first met him, for the brain is a selective organism that chooses It stores the information that is most important to us in life, and places the least important information in its hidden conventions so that it hardly appears.
Scientists have found that excessive forgetfulness, especially in small details, is evidence of the integrity of the mind, and a method that the brain uses to renew its vitality and cells and maintain its effectiveness, especially in times of stress and tension and the extreme need to make sound decisions in difficult times.
Knowing how the mind neglects unimportant information at a given moment is just as important as its ability to remember the most important information that is directly related to the topic or challenge that is being dealt with because the real goal of memory is to improve the ability to make the right decision at the right time without being overwhelmed with details.
You see: Can we teach artificial intelligence to do the same, keep the most important information and erase or put less important information in the limelight?
This is exactly what Facebook is trying to do. The world's largest social media company said it is teaching its AI systems to forget the less important information they collect, so they can complete their work more quickly and efficiently.
The network which more than 2.85 billion people enter from all over the world every month - relies heavily on artificial intelligence to detect and classify harmful content that may be transmitted through it, such as hate speech, violence, racial and racial discrimination, and other behaviors that Facebook considers dangerous and inconsistent with its policy.
In this context, scientists and researchers working for the company said - in a recently published research that they have devised a new method known as "Expire-Span" that teaches artificial intelligence how to forget large amounts of irrelevant information.
Facebook researchers Angelina Van and Sienbayer Shakhbatar and the study participants said in a joint article, “Every piece of information that is published on the network gets its own expiration date through this system, which leads to freeing up an important area of computer memory so that it can focus. On the things necessary and most important to complete the task required of him in the best way.
For example, if the model is practicing a specific task of predicting words, it can teach the AI to remember the most important words and overlook common and unrelated words that many users would use in their posts.
The article added Facebook is closer than ever to making computers retain memories, just like human brains keep the most important information and memories instead of unimportant details and memories.
The two researchers continued by saying that the scalability through this system is amazing and interesting, and we may someday be able to develop the capabilities of artificial intelligence to perform very complex operations to function exactly like the human brain.