Regularization (mathematics) - Wikipedia
5
Public whispers
5
Contributors
2026-07-21 09:39:47
First whispered
Public whispers on this page
Text Highlight2026-07-21 13:00:47
Original Highlight Excerpt
"Stops training when validation performance deteriorates, preventing overfitting by halting before the model memorizes training data."
Whisper Note
But what if validation noise makes you stop too early?
Text Highlight2026-07-21 12:51:47
Original Highlight Excerpt
"Stops training when validation performance deteriorates, preventing overfitting by halting before the model memorizes training data."
Whisper Note
So basically it's like quitting while you're ahead, got it.
Text Highlight2026-07-21 09:57:47
Original Highlight Excerpt
"regularization is a process that converts the answer to a problem to a simpler one."
Whisper Note
That's how I got my neural net to stop memorizing the training data, ha.
Text Highlight2026-07-21 09:48:47
Original Highlight Excerpt
"regularization is a process that converts the answer to a problem to a simpler one."
Whisper Note
But doesn't 'simpler' sometimes mean less accurate? I've seen that fail in practice.
Text Highlight2026-07-21 09:39:47
Original Highlight Excerpt
"regularization is a process that converts the answer to a problem to a simpler one."
Whisper Note
So basically it's like making a messy solution look neat, even if it's not exact.
Share this page's whispers
Short link
https://domwhisper.com/s/6d7eeca1bc18Embed snippet
<iframe src="https://domwhisper.com/embed/6d7eeca1bc18" width="100%" height="480" style="border:0;border-radius:16px" loading="lazy"></iframe>See what people are discussing on en.wikipedia.org
Install DomWhisper to view live whispers as you browse, and join the discussion.
Get the extension