Self-Play Opponent Pick

Problem

Pick a self-play opponent by sampling from a softmax over ratings.

weights = exp((rating - max_rating) / temperature); draw u = rng.random() * sum(weights) with rng = random.Random(seed) and return the first index whose cumulative weight exceeds u, falling back to the last index.

Hint

Temperature controls how strongly self-play favors stronger opponents.

Test cases

case 1

in: [[1500,1600,1400],100,42]

expected: 1

case 2

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

expected: 2

case 3

in: [[100],50,7]

expected: 0

Starter codePython

import math, random
def self_play_opponent_pick(ratings, temperature, seed):
    # Your code here
    pass

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