APE-X Priority Assignment

Problem

Assign a priority to a new transition the APE-X way: the new priority is max over the existing buffer priorities and (|td_error| + epsilon) ** alpha.

Hint

Giving new transitions the buffer maximum encourages them to be replayed soon.

case 1

in: [[1,2,0.5],3,1,0.000001]

expected: 3.000001

case 2

in: [[5],0.1,1,0.000001]

expected: 5

case 3

in: [[],0,0.5,0.000001]

expected: 0.001

Starter codePython

def apex_priority_assignment(existing_priorities, td_error, alpha, epsilon):
    # Your code here
    pass

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