TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Don't Cite the No Free Lunch Theorem

61 点作者 another将近 6 年前

5 条评论

jmmcd将近 6 年前
Good! This article is about the NFL for machine learning (Wolpert), and also mentions NFL for optimisation (Wolpert and Macready). My recent paper [1] is about NFL for optimisation and mentions NFL for ML. The overall message is the same: most people misunderstand it, and you probably shouldn&#x27;t cite it. And the anthropic principle is a sufficient assumption for this.<p>[1] <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1906.03280" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;abs&#x2F;1906.03280</a>
评论 #20441210 未加载
meerita将近 6 年前
The first thing that came to my mind was that Milton Friedman quote &quot;there&#x27;s no such free meal, someone has to pay it&quot;.
评论 #20440892 未加载
评论 #20441536 未加载
yters将近 6 年前
Even though the NFL only holds absolutely when problems are closed under permutation, that doesn&#x27;t imply that most problems are a good fit for a particular algorithm: <a href="https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;1609.08913.pdf" rel="nofollow">https:&#x2F;&#x2F;arxiv.org&#x2F;pdf&#x2F;1609.08913.pdf</a>
eli_gottlieb将近 6 年前
&gt;What it actually (vaguely) says is “You can’t learn from data without making assumptions”.<p>That&#x27;s what they taught us it means in our graduate machine learning class. What were people citing it to mean?
评论 #20449467 未加载
评论 #20443055 未加载
bradknowles将近 6 年前
TANSTAAFL?
评论 #20441092 未加载