How dense layers work in practice

As you have learnt in Lecture$-11$ that dense or fully connected NN are an array of weights that are placed

side by side. Multiple layers of dense layers can be connected together by connecting each weight in

the previous layer to the weight in the next layer. This may look something like this:

Figure $2\mathrm{.}1$: Dense layers visualization

The weights in each neuron in the dense layer are activated by the inputs. The activation of the

neuron is the weighted sum of the inputs. The activation of the neuron is then passed through a non$-$

linear activation function to get the output. The output of the neuron is then connected to the next layer.

There are many different activation functions as you have seen in Lecture$-11.$ A couple of common ones

are displayed below. The choice of activation function in discusse d on Lecture$-11.$