← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 8663.a · 生成模型:GPT-6

标签平滑改变监督分布

理解等级:高级

常见标签平滑把独热目标变为(1−ε)one_hot+ε/K,降低对单一类的极端监督。它不是简单让logits接近0或1,也不保证自动校准或纠正错标签;任务质量与校准都需另测。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      Label smoothing Label smoothing is a technique that not only helps with combatting overfitting but also aids in model calibration. Calibration is an underappreciated topic in deep learning. A model is said to be well-calibrated if there is a correlation between its output probability values and task accuracy. For example, consider a task that classifies a sentence as being abusive or not. If the model

    参考来源

    - 位置:《Designing Large Language Model Applications》第06章《微调》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Designing Large Language Model Applications/_epub-src/OEBPS/ch06.html) - 结论依据:摘录中关于标签平滑改变监督分布的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:Label smoothing Label smoothing is a technique that not only helps with combatting overfitting but also aids in model calibration. Calibration is an underappreciated topic in deep learning. A model is said to be well-calibrated if there is a correlation between its output probability values and task accuracy. For example, consider a task that classifies a sentence as being abusive or not. If the model