class=”sc-3778e872-0 cKDKQr”>
The new algorithm was presented to experts in the famous journal “Nature” on Wednesday.
When you watch a movie, millions of neurons fire in your brain. They respond to various features: objects, colors, textures and even emotions evoked while watching and understanding the movie. However, researchers can only examine a small part of the brain at a time.
Scientists at EPFL have now filled this void with a program called “Cebra” based on artificial intelligence. The new algorithm learns the hidden structure in the neural network. If he knows a small part of the data, he can make the remaining estimates. “Cebra” is able to show what the mouse sees with the incoming data, with less than one percent of the neurons.
For the study, the scientists had rats watch a black and white movie from the 1960s where a man ran into a car and opened the trunk. They could see the remake of the movie through the eyes of mice on a screen. The movie remade by «Cebra» almost completely matched the original. Except for minor imperfections.
About 10 years ago, it was possible to decode simple shapes from the brains of animals or humans. “But now we’re at a point where we can literally decode film footage, which we’ve never been able to do before,” said study author Mackenzie Mathis.
Scientists have also tested “Cebra” in other areas. It was also able to use brain data to predict what movement the primates would make and reconstruct the position of the mice when running through an arena.
According to Mathis, the work is a step towards the development of brain-machine interfaces based on the ability to translate brain signals into control signals for devices or machines. In the future, prosthetics, even computers and drones, must be controlled by thought.
Researchers are also seeing clinical applications. Machine learning can be used to learn the hidden structure in neural codes to better understand how the brain processes information.
(SDA)
Source : Blick
I am Dawid Malan, a news reporter for 24 Instant News. I specialize in celebrity and entertainment news, writing stories that capture the attention of readers from all walks of life. My work has been featured in some of the world’s leading publications and I am passionate about delivering quality content to my readers.
