Probabilistic neural networks can be used for classification problems. When an input is presented, the first layer computes distances from the input vector to the training input vectors and produces a vector whose elements indicate how close the input is to a training input. The second layer sums these contributions for each class of inputs to produce as its net output a vector of probabilities. Finally, a compete transfer function on the output of the second layer picks the maximum of these probabilities, and produces a 1 for that class and a 0 for the other classes. The architecture for this system is shown below.
It is assumed that there are Q input vector/target
vector pairs. Each target vector has
One of these elements is 1 and the rest are 0. Thus, each input vector
is associated with one of
The first-layer input weights, IW1,1 (
are set to the transpose of the matrix formed from the Q training
an input is presented, the
|| box produces
a vector whose elements indicate how close the input is to the vectors
of the training set. These elements are multiplied, element by element,
by the bias and sent to the
function. An input vector close to a training vector is represented
by a number close to 1 in the output vector a1.
If an input is close to several training vectors of a single class,
it is represented by several elements of a1 that
are close to 1.
The second-layer weights, LW1,2 (
are set to the matrix T of target
vectors. Each vector has a 1 only in the row associated with that
particular class of input, and 0s elsewhere. (Use function
ind2vec to create the proper vectors.)
The multiplication Ta1 sums
the elements of a1 due
to each of the
K input classes. Finally, the second-layer
a 1 corresponding to the largest element of n2,
and 0s elsewhere. Thus, the network classifies the input vector into
K class because that class has the maximum
probability of being correct.
You can use the function
create a PNN. For instance, suppose
that seven input vectors and their corresponding targets are
P = [0 0;1 1;0 3;1 4;3 1;4 1;4 3]'
P = 0 1 0 1 3 4 4 0 1 3 4 1 1 3 Tc = [1 1 2 2 3 3 3]
Tc = 1 1 2 2 3 3 3
You need a target matrix with 1s in the right places. You can
get it with the function
It gives a matrix with 0s except at the correct spots. So execute
T = ind2vec(Tc)
T = (1,1) 1 (1,2) 1 (2,3) 1 (2,4) 1 (3,5) 1 (3,6) 1 (3,7) 1
Now you can create a network and simulate it, using the input
make sure that it does produce the correct classifications. Use the
vec2ind to convert the
Y into a row
Yc to make
the classifications clear.
net = newpnn(P,T); Y = sim(net,P); Yc = vec2ind(Y)
Yc = 1 1 2 2 3 3 3
You might try classifying vectors other than those that were
used to design the network. Try to classify the vectors shown below
P2 = [1 4;0 1;5 2]' P2 = 1 0 5 4 1 2
Can you guess how these vectors will be classified? If you run the simulation and plot the vectors as before, you get
Yc = 2 1 3
These results look good, for these test vectors were quite close to members of classes 2, 1, and 3, respectively. The network has managed to generalize its operation to properly classify vectors other than those used to design the network.
You might want to try
demopnn1. It shows
how to design a PNN, and how the network can successfully classify
a vector not used in the design.