Artificial Neural Network: the description
Updated: Jan 29, 2020
Artificial Neural Network is the mimic of human nervous system.The working model of neuron is as similar to the model of Artificial Neural Network. There are several algorithms present which are helpful for the linear hypothesis problems. But Neural Network turns out to be a much better way to learn complex nonlinear hypothesis .In a neuron model number of input wires are dendrites and out out wires are axon terminal which sends message to the other neurons. Similarly the neural network has one input layer and one output layer.There are may be number of inputs present in the input layer which propagates the information to the hidden layer where all the computational work is done and finally the output layer gives the desired output.A neural net consists of a large number of simple processing elements called neurons, units, cells or nodes.
Each neuron is connected to other neurons by means of directed communication links, each with associated weight.The weight represent information being used by the net to solve a problem.
Each neuron has an internal state, called its activation or activity level, which is a function of the inputs it has received. Typically, a neuron sends its activation as a signal to several other neurons.
It is important to note that a neuron can send only one signal at a time, although that signal is broadcast to several other neurons.
There are mainly two types of neural networks are present such as convolutional neural network(CNN) and recurrent neural network(RNN).There are several advantages of RNN over FCN(Fully Connected Network).
Advantages of ANN are as follows-
1.Adaptive learning: An ability to learn how to do tasks based on the data given for training or initial experience.
2. Self-Organisation: An ANN can create its own organisation or representation of the information it receives during learning time.
3. Real Time Operation: ANN computations may be carried out in parallel, and special hardware devices are being designed and manufactured which take advantage of this capability.
4. Pattern recognition: is a powerful technique for harnessing the information in the data and generalizing about it. Neural nets learn to recognize the patterns which exist in the data set.
5. The system is developed through learning rather than programming.. Neural nets teach themselves the patterns in the data freeing the analyst for more interesting work.
Disadvantages of ANN are as follows-
•ANN is not a daily life general purpose problem solver.
• There is no structured methodology available in ANN.
• There is no single standardized paradigm for ANN development.
• The Output Quality of an ANN may be unpredictable.
• Many ANN Systems does not describe how they solve problems.
• Black box Nature
• Greater computational burden.
• Proneness to over fitting
By-Sushree Barsa Pattnayak
Intern of CoE-AI,CET BBSR