t-Closeness Check

Problem

Return True when a group's sensitive-attribute distribution is within total variation distance t of the global distribution:

TV = 0.5 * sum(|group_i - global_i|) <= t

Use a tolerance of 1e-12 so exact boundary cases still pass.

Hint

t = 0 forces every group to match the global distribution exactly.

case 1

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

expected: true

case 2

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

expected: true

case 3

in: [[1,0],[0.5,0.5],0.4]

expected: false

case 4

in: [[0.25,0.75],[0.5,0.5],0.3]

expected: true

Starter codePython

def t_closeness_holds(group_dist, global_dist, t):
    # Your code here
    pass

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