In May of this year, Google Brain researchers introduced AutoML, an artificial intelligence that generates their own artificial intelligence. AutoML now has a better artificial intelligence than human colleagues.
Terminator and Matrix are two science-fiction films that stand at a crucial point in terms of artificial intelligence prediction. As the followers know, there is a war between humans and artificial intelligence in these two films. But when you go to the bases of this war, you will see the people who created an artificial intelligence, and then the artificial intelligence that created their own artificial intelligence. The rest is what you all know; artificial intelligence that does not need people to take control of us and the beginning war. Of course, this creative artificial intelligence developed by Google is still very young for this business, but Hawking and Musk have no reason not to do what they say.
Google’s developers have automated machine learning using an approach called reinforcement learning. Thus, artificial intelligence called AutoML has become a neural network that monitors the artificial intelligence produced for a particular task. Researchers say that the task of this young artificial intelligence, called NASNet, is to describe people, cars, backpacks, and more. That’s where AutoML will evaluate NASNet’s performance and apply that knowledge to new artificial intelligence.
NASNet was highlighted by the Google-developed image systems such as ImageNet image classification and COCO object detection. According to researchers, NASNet has 82.7 percent more accurate estimates than ImageNet. This means that 1.2 percent better than previously published results, the system is 4 percent more efficient and 43.1 percent more accurate (mAP).
Machine learning gives many AI systems the ability to perform specific tasks. Although the concept behind it is quite simple, an algorithm learns by processing a ton of data. This means that the process requires a lot of time and effort. Here, Google engineers automate this situation by giving the hands of an artificial intelligence of the yarn, saving both time and labor.
It is thought that Google’s new system AutoML can be used in autonomous vehicles, robotic and many other technological areas. On the other hand, the development of systems like NASNet brings new threats. Surveillance systems, especially those seen in dystrophic science-fiction films, can provide a great improvement and individual freedom can suffer great losses.
On the other hand, Amazon, Facebook, Apple and other companies, the leading companies in this field, have established a common artificial intelligence community to develop artificial intelligence around a responsibility. However, various countries continue to do some work against the dangers of autonomous weapons.