按年被提到
20152015 · 2 次2026
机器学习与深度学习›机器学习实践与反思·arXiv:1502.04585 · cs.LG

The Ladder: A Reliable Leaderboard for Machine Learning Competitions

Avrim Blum, Moritz Hardt · 2015
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2015-06-04 · 最近 2015-12-14
2不同的人
2活跃月份
2讨论串
85被引(OpenAlex)
arXiv 摘要(候选层不写导读)

The organizer of a machine learning competition faces the problem of maintaining an accurate leaderboard that faithfully represents the quality of the best submission of each competing team. What makes this estimation problem particularly challenging is its sequential and adaptive nature. As participants are allowed to repeatedly evaluate their submissions on the leaderboard, they may begin to overfit to the holdout data that supports the leaderboard. Few theoretical results give actionable advice on how to design a reliable leaderboard. Existing approaches therefore often resort to poorly understood heuristics such as limiting the bit precision of answers and the rate of re-submission. In this work, we introduce a notion of "leaderboard accuracy" tailored to the format of a competition. We introduce a natural algorithm called "the Ladder" and demonstrate that it simultaneously supports strong theoretical guarantees in a fully adaptive model of estimation, withstands practical adversarial attacks, and achieves high utility on real submission files from an actual competition hosted by Kaggle. Notably, we are able to sidestep a powerful recent hardness result for adaptive risk estimation that rules out algorithms such as ours under a seemingly very similar notion of accuracy. On a practical note, we provide a completely parameter-free variant of our algorithm that can be deployed in a real competition with no tuning required whatsoever.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论davmre · 2015-06

“This doesn't affect the actual contest results on the test set. Anyone who wants to overfit to the validation set can do so, getting a nice temporary leaderboard spot at the price of test set generalization, which is what determines the winners. That said it'd obviously be nice to build leaderboards that are robust to this type of gaming. There's been so…”

Our perfect submission原评论 ↗
评论jmount · 2015-12

“Nice article on why it is cheating (from a mathematical sense, independent of language) to get scores from the hold-out leader board too many times (plus some methods to mitigate the effect): http://arxiv.org/abs/1502.04585.”

A Look into Machine Learning's First Cheating Scandal原评论 ↗