Short Bytes: The researchers have uncovered the algorithm that could tell us the secrets of our brain and inspire the development of true artificial intelligence. The research reveals that our brain power is based on a mathematical logic N = 2i-1. The logic governs how similar neuron groups come together and handle various tasks.
With the fast development in the field of artificial intelligence, the human intelligence is facing multiple challenges from the machines. Earlier this year, we saw how Google’s AI defeated Go champions. Apart from challenging human brains, the AI-based technologies are also changing how our computers work.
However, there remains a thick line that distinguishes how human brain works. To study this difference and incorporate the same in their products, technology companies are investing billions. But, what if I tell you that human thought and learning process can be defined using a single algorithm?
A recent research paper, published in Frontiers in Systems Neuroscience journal, suggests that a “relative simple mathematical logic underlies our complex computations”. This theory was first proposed by Joe Tsien, a researcher and neuroscientist, in October 2015.
The detailed research paper describes how groups of similar neurons (cliques) that form multiple attachments handle basic ideas of information. The researchers label these groups as “functional connectivity motifs (FCM)”, which govern the formation of all combination of ideas.
The researchers gave different food combinations to animals. Using electrodes placed at specific areas of the brain, they were able to record the response of neurons. They identified all 15 different combinations (or cliques) that responded to the food combinations.
To test their theory, the researchers studied how the algorithm works in 7 different regions of the brain that deal with handling basic tasks. They found that the number of cliques needed for an FCM is represented by a power-of-two-based permutation logic, i.e., N=2i–1.
This whole research seems pretty fascinating as it’ll help the humans to understand their own behavior. It’ll also inspire groundbreaking developments in the field of AI and create more intelligent robots.
Did you find this development in AI research interesting? Don’t forget to share your feedback in the comments section.
Research Paper: journal.frontiersin.org