← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6555.a · 生成模型:GPT-6

通用逼近不保证训练可行

理解等级:中级

某类网络具备通用函数逼近能力,是关于表示能力及相应数学条件的判断。有限数据、有限网络和优化算法不因此必然学到正确机制,也不自动保证样本外精度或因果识别。不能把“能表示”直接换成“会学到”。

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

      5.1.1 Leveraging the universal function approximator Deep learning is a highly effective universal function approximator. Let’s imagine there is a function that maps some set of inputs to some set of outputs, but we either don’t know the function or it’s too hard to write down in math or code. Given enough examples of those inputs and outputs, deep learning can approximate that function with

    参考来源

    - 位置:《Causal AI》第05章《连接因果与深度学习》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Causal AI/_epub-src/OEBPS/Text/chapter-5.xhtml) - 结论依据:摘录中关于通用逼近不保证训练可行的定义、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:5.1.1 Leveraging the universal function approximator Deep learning is a highly effective universal function approximator. Let’s imagine there is a function that maps some set of inputs to some set of outputs, but we either don’t know the function or it’s too hard to write down in math or code. Given enough examples of those inputs and outputs, deep learning can approximate that function with