Neural Network

Neural Network

1. PERCEPTRONS NEURONS

   • developed in 1950s and 1960s by scientist Frank Rosenblatt
   • Working Mechanism
     • takes several binary inputs and produces a single binary output
     • outcome is always in the form of 0 and 1
     • outcome is determined by whether the weighted sum ∑j wjxj is less than or greater than threshold value
     • varying weights and threshold can result in different models of decision making
     • output=0 if ∑j wjxj <= threshold value
     • output=1 if ∑j wjxj > threshold value

2. SIGMOID NEURONS

   • just like perceptrons, it takes several inputs but not only 0 and 1, also the values between 0 and 1 such as 0.53, 0.55....
   • it has weight for each inputs and overall bias b
   • calculated by formula, σ(w⋅x+b), where σ is called the sigmoid function
   • the output is not 0 and 1

Imágenes y archivos del mapa