Pairwise Ranking Loss

Task

Implement pairwise_ranking_loss(positive_scores: list[float], negative_scores: list[float], margin: float) -> float. Compute the mean margin-based pairwise ranking loss for aligned positive and negative scores.

Requirements

• positive_scores and negative_scores have the same shape, and each aligned pair represents one ranking comparison.
• For each pair, compute max(0, margin - positive_score + negative_score).
• Return the arithmetic mean of all per-pair losses.
• Return 0.0 when the inputs are empty.
• A pair exactly on the margin boundary contributes 0.0.
• Do not modify the inputs.

Example

positive_scores = [2.0, 1.0, 3.0]
negative_scores = [0.5, 1.5, 1.0]

pairwise_ranking_loss(positive_scores, negative_scores, margin=1.0)
# 0.5