Professional Documents
Culture Documents
Self Organizing Maps
Self Organizing Maps
output vector o
O1 O2 O3 … O
m
I1 I2 … In
input vector x
Two-dimensional
(connections omitted,
i
only neighborhood
relations shown [green])
Neighborhood of neuron i
November 24, 2009 Introduction to Cognitive Science 5
Lecture 21: Self-Organizing Maps
Self-Organizing Maps (Kohonen Maps)
A neighborhood function (i, k) indicates how closely
neurons i and k in the output layer are connected to
each other.
Usually, a Gaussian function on the distance between
the two neurons in the layer is used:
position of k
November 24, 2009
position of i
Introduction to Cognitive Science 6
Lecture 21: Self-Organizing Maps
Unsupervised Learning in SOMs
For n-dimensional input space and m output neurons:
0 20 100
X1 X2 … XN