"A Child's Eyes" teaches artificial intelligence how humans evolve! "A Child's Eyes" teaches artificial intelligence how humans evolve!

"A Child's Eyes" teaches artificial intelligence how humans evolve!

"A Child's Eyes" teaches artificial intelligence how humans evolve!

Scientists have trained artificial intelligence to learn about human evolution through the eyes of a child, amid fears that modern technology could destroy us.
The New York University research team linked the head camera recorder to baby Sam when he was just six months old until his second birthday.

The AI ​​model was fed snapshots consisting of 250,000 words as well as a set of images, where it learned how to recognize different objects similar to what Sam did.

The AI ​​developed its knowledge as a child did, by observing the environment, listening to people nearby, and connecting the dots between what was seen and heard.

The experiment also identified the relationship between visual and linguistic representation in child development.

Scientists set out to discover how humans associate words with visual representations, such as associating the word “ball” with a round object rather than other features, objects, or events.

The camera randomly captured Sam's daily activities, such as meal times, reading books, and the child's play, amounting to about 60 hours of data.

The team used a vision and text encoder to translate images and written language for the AI ​​model, from footage obtained through the SAM headset.

Although the snapshots often did not directly link words and images, the Child Vision Variant Learning (CVCL) robot, which consists of artificial intelligence and a front-facing camera, was able to recognize the meanings.

The model used a differentiated learning approach that builds information to predict which images and text go together.

The team gave several tests to 22 separate words and pictures that were present in the child's video footage, and found that the model was able to correctly match many of the words and their pictures.

The results showed that the AI ​​model could generalize what it learned with an accuracy rate of 61.6%.

The scientists found that there were still flaws in the artificial intelligence model, and while the test showed promise in understanding how children develop cognitive functions, it was limited by its inability to fully experience a child's life.

They plan to conduct additional research to replicate early language learning in children as young as about 2 years old.

China launches the world's first satellite to test 6G technology into orbit

China on Saturday launched two experimental satellites to explore integrated space and terrestrial communications technologies. One of them is dedicated to testing 6G technology.
The satellites successfully reached low Earth orbit at an altitude of 500 km.

One of these satellites is China Mobile 01, which was jointly developed by state-owned telecommunications company China Mobile and Chinese satellite internet company UBINEXUS. It is the world's first signal processing satellite equipped with a fifth generation (5G) space-based ground operating system.

By taking advantage of the satellite's wide coverage, the base station it carries can transmit 5G signals to places that cannot be covered by terrestrial networks.

According to China Mobile, the second test satellite, called Xinghe or Star Core, features a distributed autonomous architecture for sixth generation (6G) technology, which the telecommunications company developed in cooperation with the Small Satellite Innovation Academy of the Chinese Academy of Sciences.

The Chinese telecommunications company says that Star Core is the first satellite to test sixth generation (6G) technology in the world.

China Mobile explained that compared to high-orbit satellites that operate at an altitude of 36,000 km, Star Core technology provides the advantages of low latency (communications) and high data transfer rates at an altitude of 500 km.

The company pointed out that low-Earth orbit satellites are important platforms for integrated terrestrial satellite networks in the future.

The development of these wide-range, low-orbit satellites can help compensate for signal coverage gaps in terrestrial mobile networks and extend satellite Internet services with higher bandwidth to remote areas, oceans and the aviation sector.

China Mobile confirmed that it will conduct in-orbit tests based on experimental satellites to further accelerate the development of Earth satellite technologies.
Previous Post Next Post