Minimum Expected Cost

Problem

Compute the minimum total expected cost over a list of probability predictions. For each p, the expected cost of predicting 0 is p*C[1][0] + (1-p)*C[0][0] and of predicting 1 is p*C[1][1] + (1-p)*C[0][1]; sum the smaller of the two. Empty input returns 0.0.

Hint

The optimal policy picks whichever label minimizes expected cost for each sample independently.

Test cases

case 1

in: [[0.2,0.5,0.8],[[0,1],[1,0]]]

expected: 0.8999999999999999

case 2

in: [[0.2,0.8],[[0,5],[1,0]]]

expected: 1

case 3

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

expected: 0

case 4

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

expected: 0.5

Starter codePython

def minimum_expected_cost(probs, cost_matrix):
    # Your code here
    pass

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