What is ANN? and Difference between ANN and BNN

ANN-Artificial Neural Networks

Topics to be covered:

In this topic we going to get a brief idea about ANN-

            1.1 what is ANN?

            1.2 Difference between ANN and BNN.

            1.3 Typical classes of Network Architecture.

1.1 What is ANN?

ANN is an efficient computing system whose central theme is borrowed from the analogy of the biological Neural Networks (BNN). ANN's are also named as "Artificial Neural System" (or) "Parallel Distributed processing System".

• ANN acquires a large collection of units that are interconnected in some pattern to allow communication between the units these units, also referred to as Nodes (or) Neurons, are simple processors which operating parallel.
• Every Neuron connected with other Neuron through a connection link. Each connection link is associated with a weight that information about the input signal. This is the most useful information for Neurons to solve a particular problem because the weight usually exited (or)  inhibits the signal that is being communication.
• Each Neuron has a internal state which is called an activation signal.
• Output signals are produced after combining the input signals and Activation Rule.

                      

1.2 Difference between ANN and BNN

ANN	BNN
1.ANN’s are fast in processing information.	1.BNN is slow in processing information.
2.ANN operates on the sequential modes i.e,one instruction after the another.	2.BNN can perform massively parallel operations.
3.ANN are not fault tolerant, since corrupted information cannot be retrieved from the memory.	3.They exhibit fault tolerance since the information is distributed in connection to the Network.
4.There is control unit which monitors all the activities of computing.	4.There is no specific control mechanism.
5.	5.
6.We have nodes for processing information.	6.We have cell body for processing information.
7.We have input unit to give input.	7.We have dendrites as input.
8.We have interconnection ways.	8.We have Synapse.
9.We have output.	9.We have Axons.

1.3 Typical Classes of Network Architecture

There exist five different types of neuron collection of architecture. They are

• Single Layer Feed Forward Neural Network
• Multi Layer Feed Forward Neural Network
• Single Node with its own Feedback
• Single Layer Recurrent Networks
• Multi Layer Recurrent Networks 

Single Layer Feed Forward Neural Network:

In this type of network we have two layers 

• Input layer 
• Output Layer 

But input layer does not count because no computation is performed in this layer. Output layer is found when different weights are applied on input nodes and cumulative effect per node is taken. After these the Neurons collectively give the output layer to compute the output signals.

Fig: Single layer feed forward Neural Network

Multi Layer Feed Forward Neural Network:

In this type of neural network we have three layers they are 

• Input Layer 
• Hidden Layer 
• Output Layer

In this Network hidden layer which is internal to the network and has no direct contact with external layer. Existence of one (or) more hidden layers enable the network to the computationally stronger because information flows through the input function and the intermediate computation used to be output 'Y'. 

Fig: Multi layer feed forward Neural Network

Single Node with its Own Feed-Back:

When outputs are directed back as inputs to the same layer (or) preceding layer nodes then it results in feedback Networks.

Fig: Single Node with its own feedback

Single Layer Recurrent Network: 

The above Network is single layer network with feed back connection in which processing elements output can be directed back to itself (or) to other processing elements (or) both. Recurrent Neural Network(RNN) is a class of ANN where connection between nodes from a graph along a sequences. This allows it to exhibit dynamic temporal behavior for a time sequence. Unlike Feed Forward Neural Network(FFNN), Recurrent Neural Network can use their internal state to process sequence of inputs.

Fig: Single Layer feed forward Neural Network

Multi Layer Feed Forward Neural Network:

In this type of Network, processing element output can be directed to the processing element in the same layer and in the preceding layer forming a multi layer recurrent network. They perform the same task for every element of the sequence with the output dependent on previous computation.

Fig: Multi Layer Feed Forward Neural Network