← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9162 · 生成模型:GPT-6

特征压缩应保留任务信息

理解等级:中级

特征提取把大量原始观测转为与任务有关的表示,例如线的位置、目标类别或局部描述子。输出字节更少不是唯一目标,错误删掉判别信息会损害任务;Ratslife六位标签只是已知类别集合的编码例子,不是图像所有信息只有六位。

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

      The information generated by sensors can be quite voluminous. For example, a simple webcam generates 640 × 480 color pixels (red, green, and blue) or 921,600 bytes around 30 times per second. A single-ray laser scanner provides around 600 distance measurements 10 times per second in the form of a point cloud. Consider for a moment the information that a robot requires in order to

    参考来源

    - 位置:《Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)》第09章《特征提取》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Introduction to Autonomous Robots Mechanisms, Sensors, Actuators, and Algorithms (Nikolaus Correll, Bradley Hayes etc.)/_epub-src/OEBPS/xhtml/chapter_9.xhtml) - 结论依据:摘录中关于特征压缩应保留任务信息的定义、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:The information generated by sensors can be quite voluminous. For example, a simple webcam generates 640 × 480 color pixels (red, green, and blue) or 921,600 bytes around 30 times per second. A single-ray laser scanner provides around 600 distance measurements 10 times per second in the form of a point cloud. Consider for a moment the information that a robot requires in order to