machine-learning-foundations-1

Posted on | Post modified:

Abstract: 机器学习基石（林轩田）第一讲，介绍相关基本概念。

  1. 关于学习
    learning
    $observations \rightarrow learning \rightarrow skill$
    machine learning
    $data \rightarrow machine\ learning \rightarrow skill$
    其中，
    $skill \Leftrightarrow improve\ some\ performance\ measure$

  2. 使用机器学习的关键前提：

    • exists some ‘underlying pattern’ to be learned
      ——so ‘performance measure’ can be improved
    • but no programmable (easy) definition
      ——so ‘ML’ is needed
    • somehow there is data about the pattern
      ——so ML has some ‘inputs’ to learn from

  3. 学习问题的形式化表示(formalization)

    $$learning\ model=A\ and\ H$$

  4. 机器学习与其他领域的关系
    1) definition

    - Machine Learning: use data to compute hypothesis g that approximates target f
- Data Mining: use (huge) data to find property that is interesting
- Artificial Intelligence: compute something that shows intelligent behavior
- Statistics: use data to make inference about an unknown process

    2) relationship

    - ML can help DM, and vice versa
- ML is one possible route to realize AI
- Statistics provides many tools for ML