EM Mixture Weight Update

Problem

In the M-step of EM, each mixture weight becomes the average responsibility of its component across the data set: (1 / n) times the sum over points of responsibilities[n][k]. Return the list of updated weights, or an empty list for inconsistent shapes.

Hint

Average the responsibility columns over the data points.

Test cases

case 1

in: [[[0.9,0.1],[0.8,0.2],[0.2,0.8],[0.1,0.9]]]

expected: [0.5,0.5]

case 2

in: [[[0.6,0.3,0.1],[0.2,0.5,0.3]]]

expected: [0.4,0.4,0.2]

case 3

in: [[[1],[1]]]

expected: [1]

case 4

in: [[]]

expected: []

case 5

in: [[[1,0],[0.5,0.5]]]

expected: [0.75,0.25]

Starter codePython

def em_mixture_weight_update(responsibilities):
    # Your code here
    pass

5 test cases · run your code against them in the editor