Ensemble learning - Wikipedia

公开标注数

参与人数

2026-07-23 09:41:25

首次 Whisper

本页的公开 Whisper

划选高亮2026-07-23 13:02:25

原文高亮摘录

“Empirically, ensembles tend to yield better results when there is a significant diversity among the models.”

Whisper 随想笔记

But wait, doesn't that contradict using strong models? I thought you'd want the best ones.

划选高亮2026-07-23 12:53:25

原文高亮摘录

“Empirically, ensembles tend to yield better results when there is a significant diversity among the models.”

Whisper 随想笔记

So basically, disagreement is good — like a team where everyone thinks the same is useless.

划选高亮2026-07-23 09:59:25

原文高亮摘录

“ensemble methods use multiple learning algorithms to obtain better predictive performance”

Whisper 随想笔记

Isn't this just voting but for algorithms? Seems obvious when you say it.

划选高亮2026-07-23 09:50:25

原文高亮摘录

“ensemble methods use multiple learning algorithms to obtain better predictive performance”

Whisper 随想笔记

I tried this once with random forests and it did beat my single tree model.

划选高亮2026-07-23 09:41:25

原文高亮摘录

“ensemble methods use multiple learning algorithms to obtain better predictive performance”

Whisper 随想笔记

So basically more models = better results, but that's not always true in practice.

短链接

https://domwhisper.com/s/798173b4271f

嵌入代码

<iframe src="https://domwhisper.com/embed/798173b4271f" width="100%" height="480" style="border:0;border-radius:16px" loading="lazy"></iframe>

安装 DomWhisper，浏览网页时实时查看 whisper，也可以加入讨论。