知识卡片
学习改变寻找映射的方式
视觉算法可以由人直接编写、从世界模型推导,也可用输入输出实例拟合函数。学习主要改变如何找到映射,不自动改变任务是什么或输出该表示什么。图中把SGD比作编译器是教学比喻,训练不是保证产生正确程序的机械编译。
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
- · 视觉的挑战
原书出处 · 1 处
- · 查看原书依据
Learning derives vision algorithms by fitting functions to data. One way to think about it is that our goal is to “program” a vision algorithm and there are a few ways to do so (figure 1.15). One way is to write the algorithm directly in Python code: first we will compare the intensity of this pixel and that pixel, then we will threshold about some
参考来源
- 位置:《Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)》第01章《视觉的挑战》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Foundations of Computer Vision (Antonio Torralba, Phillip Isola etc.)/_epub-src/OPS/xhtml/c001.xhtml)
- 结论依据:摘录中关于学习改变寻找映射的方式的定义、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:Learning derives vision algorithms by fitting functions to data. One way to think about it is that our goal is to “program” a vision algorithm and there are a few ways to do so (figure 1.15). One way is to write the algorithm directly in Python code: first we will compare the intensity of this pixel and that pixel, then we will threshold about some