Dueling Centered Advantage

Problem

Center a dueling advantage stream by subtracting its mean.

Return advantage[a] - mean(advantage) for every action, which makes the advantage identifiable without changing the greedy action. An empty list returns [].

Hint

The value stream must carry the average action quality once advantages are centered.

Test cases

case 1

in: [[1,2,3]]

expected: [-1,0,1]

case 2

in: [[5,5]]

expected: [0,0]

case 3

in: [[]]

expected: []

case 4

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

expected: [-1,0,1]

Starter codePython

def dueling_centered_advantage(advantage):
    # Your code here
    pass

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