Gradient boosting - Wikipedia
6
公开标注数
6
参与人数
2026-07-20 10:03:22
首次 Whisper
本页的公开 Whisper
划选高亮2026-07-20 16:27:22
原文高亮摘录
“Gradient boosting is typically used with decision trees (especially CARTs) of a fixed size as base learners.”
Whisper 随想笔记
Fixed-size trees keep it simple but I wonder if that limits flexibility.
划选高亮2026-07-20 13:24:22
原文高亮摘录
“The idea of gradient boosting originated in the observation by Leo Breiman that boosting can be interpreted as an optimization algorithm on a suitable cost…”
Whisper 随想笔记
So Friedman and Mason were working on this around the same time? Wild.
划选高亮2026-07-20 13:15:22
原文高亮摘录
“The idea of gradient boosting originated in the observation by Leo Breiman that boosting can be interpreted as an optimization algorithm on a suitable cost…”
Whisper 随想笔记
Breiman really nailed it—boosting as optimization just clicks.
划选高亮2026-07-20 10:21:22
原文高亮摘录
“Gradient boosting is a machine learning technique based on boosting in a functional space, where the target is pseudo-residuals instead of residuals as in traditional…”
Whisper 随想笔记
I've seen gradient-boosted trees beat random forest in practice, no contest.
划选高亮2026-07-20 10:12:22
原文高亮摘录
“Gradient boosting is a machine learning technique based on boosting in a functional space, where the target is pseudo-residuals instead of residuals as in traditional…”
Whisper 随想笔记
Does this mean gradient boosting is just a fancy term for tweaking residuals?
划选高亮2026-07-20 10:03:22
原文高亮摘录
“Gradient boosting is a machine learning technique based on boosting in a functional space, where the target is pseudo-residuals instead of residuals as in traditional…”
Whisper 随想笔记
So it's basically fitting trees to the mistakes, but in a smarter way than old boosting.
分享本页 Whisper
短链接
https://domwhisper.com/s/36477a14dfb2嵌入代码
<iframe src="https://domwhisper.com/embed/36477a14dfb2" width="100%" height="480" style="border:0;border-radius:16px" loading="lazy"></iframe>