Neural networks types include perceptrons, Hopfield networks, Boltzmann machines, fully connected neural networks, convolutional neural networks, re-current neural networks, long/short term memory neural networks, auto-encoders, deep belief networks, generative adversarial networks. Majority of them are trained with an algorithm called back-propagation.

Convolutional neural networks which could be considered essentially a not fully connected neural nets in which each neuron is connected to only a few neurons in the previous layer and neurons share weights. These type of networks have been proven successfully especially in the fields of computer vision and natural language processing, where they broke every record.

Thanks a lot.

Dear Dr Cifci,

I entirely agree with Dr Naderpour's answer. I have also attached a link with a tutorial explaining convolutional neural networks:

http://ufldl.stanford.edu/tutorial/supervised/ConvolutionalNeuralNetwork/

In terms of your other question (PNN), maybe the following link is helpful: https://www.cse.unr.edu/~looney/cs773b/PNNtutorial.pdf

Yours sincerely,

Rainer

Dr Rainer Spiegel, your answer was perfect magestic and fully democratic or universal.

Thanks for your example of correctness universality and open heart.

RCB