<< Chapter < Page
  Machine learning   Page 1 / 1
Chapter >> Page >

The k -means clustering algorithm

In the clustering problem, we are given a training set { x ( 1 ) , ... , x ( m ) } , and want to group the data into a few cohesive “clusters.”Here, x ( i ) R n as usual; but no labels y ( i ) are given. So, this is an unsupervised learning problem.

The k -means clustering algorithm is as follows:

  • Initialize cluster centroids μ 1 , μ 2 , ... , μ k R n randomly.
  • Repeat until convergence: {
    • For every i , set
      c ( i ) : = arg min j | | x ( i ) - μ j | | 2 .
    • For each j , set
      μ j : = i = 1 m 1 { c ( i ) = j } x ( i ) i = 1 m 1 { c ( i ) = j } .
    • }

In the algorithm above, k (a parameter of the algorithm) is the number of clusters we want to find; and the cluster centroids μ j represent our current guesses for the positions of the centers of the clusters. To initialize thecluster centroids (in step 1 of the algorithm above), we could choose k training examples randomly, and set the cluster centroids to be equal to the values ofthese k examples. (Other initialization methods are also possible.)

The inner-loop of the algorithm repeatedly carries out two steps: (i) “Assigning” each training example x ( i ) to the closest cluster centroid μ j , and (ii) Moving each cluster centroid μ j to the mean of the points assigned to it. [link] shows an illustration of running k -means.

K-means algorithm. Training examples are shown as dots, and cluster centroids are shown as crosses. (a) Original dataset. (b) Random initial cluster centroids (in this instance, not chosen to be equal to two training examples). (c-f) Illustration of running two iterations of k-means
K-means algorithm. Training examples are shown as dots, and cluster centroids are shown as crosses. (a) Original dataset. (b) Random initial clustercentroids (in this instance, not chosen to be equal to two training examples). (c-f) Illustration of running two iterations of k -means. In each iteration, we assign each training example to the closest cluster centroid (shown by “painting” the training examplesthe same color as the cluster centroid to which is assigned); then we move each cluster centroid to the mean of the points assigned to it. (Best viewed in color.)Images courtesy Michael Jordan.

Is the k -means algorithm guaranteed to converge? Yes it is, in a certain sense. In particular, let us define the distortion function to be:

J ( c , μ ) = i = 1 m | | x ( i ) - μ c ( i ) | | 2

Thus, J measures the sum of squared distances between each training example x ( i ) and the cluster centroid μ c ( i ) to which it has been assigned. It can be shown that k -means is exactly coordinate descent on J . Specifically, the inner-loop of k -means repeatedly minimizes J with respect to c while holding μ fixed, and then minimizes J with respect to μ while holding c fixed. Thus, J must monotonically decrease, and the value of J must converge. (Usually, this implies that c and μ will converge too. In theory, it is possible for k -means to oscillate between a few different clusterings—i.e., a few different values for c and/or μ —that have exactly the same value of J , but this almost never happens in practice.)

The distortion function J is a non-convex function, and so coordinate descent on J is not guaranteed to converge to the global minimum. In other words, k -means can be susceptible to local optima. Very often k -means will work fine and come up with very good clusterings despite this. But if you are worried about getting stuck in badlocal minima, one common thing to do is run k -means many times (using different random initial values for the cluster centroids μ j ). Then, out of all the different clusterings found, pick the one that gives the lowest distortion J ( c , μ ) .

Get Jobilize Job Search Mobile App in your pocket Now!

Get it on Google Play Download on the App Store Now




Source:  OpenStax, Machine learning. OpenStax CNX. Oct 14, 2013 Download for free at http://cnx.org/content/col11500/1.4
Google Play and the Google Play logo are trademarks of Google Inc.

Notification Switch

Would you like to follow the 'Machine learning' conversation and receive update notifications?

Ask