Neural Networks Taken Apart: Carbon vs Silicon / Man vs Machine Pt. 1

So, assuming that you’ve understood why we modeled Neural Networks after our brains, I’ll now dive into explaining and comparing the two things, to give you a better understanding of how neural networks work without too much math. Here’s how a neuron looks..

Unless you happen to have a degree in neuroscience or something, chances are that thing made your head spin. So for those of us, barely capable of telling an artery from a vein, here’s a simpler version…

Now here’s a comparison with what a neuron in a neural network looks like…

Here’s how it was simplified, the orange circle is where the magic happens, it can be called the ‘nucleus’. The black lines on the left hand side are what were called ‘dendrites’. They’re basically branches, that carry the input/information into the neuron. The ‘axon terminal’ is the black line near the right hand side, which carries the output signal away from the neuron to the next one.

Collectively, a neural network(Biological one) looks like…

Quite a mess right ? Well here’s how a neural network used in machine learning looks like…

Now, the artificial one looks a lot less daunting when compared to it’s inspiration right ? We just broke down all the complexities of neurons in our brain into a simple ball and stick model. Keep all these pictures in mind, as in the next post I’ll get into how this stuff actually works, again with as less math as possible.

Originally published at sushritpasupuleti.blogspot.com.