When asked to classify odors, artificial neural networks
adopt a structure that closely resembles that of the brain’s olfactory circuitry.
In just a few minutes, a computer model can teach itself to smell using machine learning.
Researchers discovered that when it does, it creates a neural network that closely resembles
the olfactory circuits used by animal brains to analyse odours.
From fruit flies to humans, olfactory information is processed in the brain in essentially the same way. However, neuroscientists who trained an artificial neural network to do a simple odour classification job were shocked to see how well it mimicked biology's technique.
Guangyu Robert Yang, an associate scientist at MIT's McGovern Institute for Brain Research who conducted the work as a postdoc at Columbia University, adds, "The method we utilise bears no relation to the actual process of evolution." The similarities between the artificial and biological systems show that the olfactory network in the brain is well-suited to its function.
Now we'll talk about how artificial neural networks can learn to smell like a brain.
www.dprg.co.in
