Presentation is loading. Please wait.

Presentation is loading. Please wait.

 How many neurons are to be used?  How the neurons are to be connected to form a network.  Which learning algorithm to use?  How to train the neural.

Similar presentations


Presentation on theme: " How many neurons are to be used?  How the neurons are to be connected to form a network.  Which learning algorithm to use?  How to train the neural."— Presentation transcript:

1

2  How many neurons are to be used?  How the neurons are to be connected to form a network.  Which learning algorithm to use?  How to train the neural network?  Training: Initialize the weights of the network and update the weights from a set of training examples

3 Neuron w1w1 w2w2 wnwn Y...... x1x1 x2x2 x3x3 Input signals WeightOutput Signals

4  The neuron computes the weighted sum of the input signals and compares the result with a threshold value of, T h  If the net weighted input is less than the threshold the neuron output is –1.  If the net weighted input is greater than or equal to the threshold, the neuron becomes activated and its output attains a value +1 (This type of activation function is called a sign function)

5 w1 w2 Σ Th Threshold Linear Combiner Hard Limiter Y-output x1 x2  Step & sign activation function called hard limit functions.  Single neuron with adjustable synaptic weight and a hard limiter.

6  Step 1: Initialization Set the initial weights w 1,w 2,….w n and Threshold-Th  Step 2: Activation Active the perceptron by applying inputs x 1 (p), x 2 (p)….. x n (p) and desired output Y d (p). Where p iteration, n number of inputs  Step 3: Weight Training Update the weight of the perceptron.  Step 4: Iteration Increase iteration p by one, go back to step 2 and repeat the process.

7 EpochInputsY(d)Initial WeightOutput Y Error e Final Weight x1x2Ydw1w2w1w2 10000.3-0.1000.3-0.1 0100.3-0.1000.3-0.1 1000.3-0.110.2-0.1 1110.2-0.1010.30.0 20000.30.0000.30.0 0100.30.0000.30.0 1000.30.010.20.0 1110.20.0100.20.0 Threshold=0.2, Learning rate = 0.1

8 EpochInputsY(d)Initial WeightOutput Y ErrorFinal Weight x1x2Ydw1w2w1w2 30000.20.0000.20.0 0100.20.0000.20.0 1000.20.010.10.0 1110.10.0010.20.1 40000.20.1000.20.1 0100.20.1000.20.1 1000.20.110.1 111 10 Threshold=0.2, Learning rate = 0.1

9 EpochInputsY(d)Initial WeightOutput Y ErrorFinal Weight x1x2Ydw1W2w1w2 50000.1 00 010 00 100 00 111 10 Threshold=0.2, Learning rate = 0.1

10  [Negnevitsky, 2001] M. Negnevitsky “ Artificial Intelligence: A guide to Intelligent Systems”, Pearson Education Limited, England, 2002.  [Russel, 2003] S. Russell and P. Norvig Artificial Intelligence: A Modern Approach Prentice Hall, 2003, Second Edition  [Patterson, 1990] D. W. Patterson, “Introduction to Artificial Intelligence and Expert Systems”, Prentice-Hall Inc., Englewood Cliffs, N.J, USA, 1990.  [Minsky, 1974] M. Minsky “A Framework for Representing Knowledge”, MIT-AI Laboratory Memo 306, 1974.


Download ppt " How many neurons are to be used?  How the neurons are to be connected to form a network.  Which learning algorithm to use?  How to train the neural."

Similar presentations


Ads by Google