Artificial Neural Network

 Artificial neural nerwork (ANN) is an efficient information processing system which resembles in characteristics with a biological neural nerwork. ANNs possess large number of highly interconnected processing elements called notUs or units or neurom, which usually operate in parallel and are configured in regular architectures. 

Each neuron is connected wirh the oilier by a connection link. Each connection link is associated with weights which contain info!£11ation about the_iapu.t signal. This information is used by rhe neuron n;t to solve a .Particular pr.cl>lem. ANNs' collective behavior is characterized by their ability to learn, recall and' generaUa uaining p®:erns or data similar to that of a human brain. They have the capability ro model networkS of ongma:l nellfOIIS as-found in the brain. Thus, rhe ANN processing elements are called neurons or artificial neuro'f·

It should be noted that each neuron has an imernal stare of its own. This imernal stare is called ilie activation or activity of neuron, which is the function of the. inputs the neuron receives. The activation signal of a neuron is transmitted to other neurons. Remembe(i neuron can send only one signal at a rime, which can be transmirred to several ocher neurons. To depict rhe basic operation of a neural net, ·consider a set of neurons, say X1 and Xz, transmitting signals to a110ilier neuron, Y. Here X, and X2 are input neurons, which transmit signals, andY is the output neuron, which receives signals. Input neurons X, and Xz are connected to the output neuron Y, over a weighted interconnection links (W, and W2) 

 For the above simple rleuron net architecture, the net input has to be calculated in the following way: 

                                                    Yin= XIWI +.X2W2

where Xi and X2 of the input neurons X, and X2, i.e., the output of input signals. The output y of the output neuron Y can be o[)i"alneaOy applymg the ner input, i.e., the function of the net input:

                                                           J = f(y;,)

                                        Output= Function (net input calculated)

 The function robe applied over the l]£t input is call: ACTIVATION FUNCTION. There are various activation functions, which will be discussed in the forthcoming sect10 _ . e a ave calculation of the net input is similar tq the calculation of output of a pure linear straight line equation (y = mx). The neural net of a pure linear 

 Here, m oblain the output y, the slope m is directly multiplied with the input signal. This is a linear equation. Thus, when slope and input are linearly varied, the output is also linearly varied,  This shows that the weight involved in dte ANN is equivalent to the slope of the linear straight line.