Paired Bootstrap Win Rate
~12 mincode completion
Implement paired_bootstrap_win_rate(scores_a, scores_b, resamples) returning the fraction of resamples in which A's mean score is strictly greater than B's.
Where this shows up
~12 min
••••••••••••••••
8 employers weight this skill
4 quant funds, 2 health and bio companies, 2 big tech firms. Top match scores 93.
Python
import numpy as np
def paired_bootstrap_win_rate(scores_a, scores_b, resamples):
"""
Fraction of paired bootstrap resamples in which system A beats system B.
Args:
scores_a: (n,) per-example scores of system A
scores_b: (n,) per-example scores of system B
resamples: (R, n) integer example indices, one row per resample
Returns:
float in [0, 1]
"""
# YOUR CODE HERE
pass