← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6686.e · 生成模型:GPT-6

偏置允许不经过原点的阈值

理解等级:中级

偏置b独立于输入加入加权和,可等价为额外固定输入1及可训练权重。它让零输入也能对应不同输出和阈值位置;不设偏置时z为0,但逻辑激活输出为0.5,不能笼统说所有神经元零输入都输出0。

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

      Why the extra weight? Did you notice that you have one additional weight, w0? There is no input labeled x0, so why is there a w0? Can you guess why we always give our neural neurons an input signal with a constant value of 1.0 for x0? Think back to the linear and logistic regression models you have built in the past. Do you remember

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第05章《神经网络基础》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-5.html) - 结论依据:摘录中关于偏置允许不经过原点的阈值的定义、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:Why the extra weight? Did you notice that you have one additional weight, w0? There is no input labeled x0, so why is there a w0? Can you guess why we always give our neural neurons an input signal with a constant value of 1.0 for x0? Think back to the linear and logistic regression models you have built in the past. Do you remember