I have seen this definition of a single Neuron everywhere, that a Nueron takes some input/s (numbers) and give some output/s (numbers) but am still not able to get what it is made of?
Is it a function? What is the formula then? Is it a gate like AND, OR?
There should be some Algorithm behind it?
week-1 Advanced Learning Algorithms Neural Networks and Deep Learning AI Discussions #Neuron
Take a look at slides C2_W1.pdf slide 10. Thinking in terms of algorithms, we first compute:
z = f(x) = wx + b
This is a linear combination of the inputs x, weights w and bias b. This part represents the input being processed by the neuron, before applying any activation.
Next, we pass z through an activation function g(z), which can vary depending on the type of neuron. For example, if the activation function is linear, g(z) = z. If it’s the sigmoid function it’s 1 //(1+e^-z).
A single neuron is more than a simple AND/OR gate because the activation function introduces complexity. Well, that’s my basic understanding anyway … . I hope it’s reasonably accurate. Good luck.