Explainable artificial intelligence - Wikipedia
5
公开标注数
5
参与人数
2026-07-26 10:02:43
首次 Whisper
本页的公开 Whisper
划选高亮2026-07-26 13:23:43
原文高亮摘录
“SHAP (SHapley Additive exPlanations) enables visualization of the contribution of each input feature to the output.”
Whisper 随想笔记
But doesn't it get super slow with many features?
划选高亮2026-07-26 13:14:43
原文高亮摘录
“SHAP (SHapley Additive exPlanations) enables visualization of the contribution of each input feature to the output.”
Whisper 随想笔记
So SHAP is basically game theory for ML, cool.
划选高亮2026-07-26 10:20:43
原文高亮摘录
“A model is transparent "if the processes that extract model parameters from training data and generate labels from testing data can be described and motivated…”
Whisper 随想笔记
I think this is spot on — transparency is about the designer's ability to explain, not just the code.
划选高亮2026-07-26 10:11:43
原文高亮摘录
“A model is transparent "if the processes that extract model parameters from training data and generate labels from testing data can be described and motivated…”
Whisper 随想笔记
So basically if the guy who built it can talk about it, it's transparent? What about everyone else?
划选高亮2026-07-26 10:02:43
原文高亮摘录
“A model is transparent "if the processes that extract model parameters from training data and generate labels from testing data can be described and motivated…”
Whisper 随想笔记
That definition sounds nice but it's like saying a recipe is transparent if the chef can explain it, not if you can
分享本页 Whisper
短链接
https://domwhisper.com/s/41712b8d1788嵌入代码
<iframe src="https://domwhisper.com/embed/41712b8d1788" width="100%" height="480" style="border:0;border-radius:16px" loading="lazy"></iframe>