Working of human brain has been a riddle to the scientists and researchers all over the world have tried to unravel the mystery of brain. Although they have found success to some extent but with all their efforts they only know about 20 percent of our brain. Adding to it, a team of researchers claims that they have decoded brain science and they know how our brain manages to store such a vast pool of information and how it extracts the exact data when required.
Only 0.15 percent of the original information is required by the brain to accurately predict and categorize the data. This is why our brain works so efficiently and manages to quickly retrieve even very faded information. What’s more striking is that researchers have designed an algorithm that mimics the functionality of a human brain. Scientists applied artificial intelligence and machine learning due to which the highly advanced algorithm works just like our brain and performs equally well.
While explaining researchers from the Georgia Institute of Technology said that there’s no machine that learns more efficiently and quickly than our brain. We regularly change our style but still our brain recognize faces of dear ones with an ease that shows how well our brain performs.
Study authors conducted random projection test to test the learning ability of the brain. The test was based on showing certain images and later recognizing them with very small portion of information. For the study, every participant was shown 16 images with 150×150 pixel size for 10 seconds each. Later, scientists asked every participant to identify the image by just seeing a small part of the image.
After analyzing the data, study authors found that only 0.15 percent of the original information is required for the brain to accurately predict and categorize the right data.
After the tests, researchers developed an algorithm to mimic the learning ability of brain. After several tweaks, the algorithm based on neural network performed as well as human brain.
“We were surprised by how close the performance was between extremely simple neural networks and humans,” said Santosh Vempala, from the Georgia Institute of Technology.
“This fascinating paper introduces a localized random projection that compresses images while still making it possible for humans and machines to distinguish broad categories,” said Sanjoy Dasgupta, professor at the University of California San Diego.
The study was published in the journal Neural Computation.