知识卡片
仿射加逻辑激活对应二类模型
单个单元计算z=w·x+b,再取σ(z)=1/(1+exp(−z)),其前向函数形式对应二类逻辑回归。训练时的损失、正则、权重和优化器仍可能不同,因此只凭架构同形不能保证两种实现参数和预测完全一致。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 神经网络基础
原书出处 · 1 处
- · 查看原书依据
One of the most complex things neurons do is process language. Think about how a perceptron might be used to process natural language text. Does the math shown in figure 5.2 remind you of any of the machine learning models you’ve used before? What machine learning models do you know of that multiply the input features by a vector of weights or coefficients? Well, that
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:摘录中关于仿射加逻辑激活对应二类模型的定义、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:One of the most complex things neurons do is process language. Think about how a perceptron might be used to process natural language text. Does the math shown in figure 5.2 remind you of any of the machine learning models you’ve used before? What machine learning models do you know of that multiply the input features by a vector of weights or coefficients? Well, that