# Machine Learning Foundations note No.01

## February 27, 2014

### 前言

 1 机器学习基石 
 这门课程的相关内容,在这里留下一些记录

### 第一课: The Learning Problem

• When Can Machines Learn? (illustrative + technical)
• Why Can Machines Learn? (theoretical + illustrative)
• How Can Machines Learn? (technical + practical)
• How Can Machines Learn Better? (practical + theoretical)

Google Doc上第一课的PPT • input: $x \in X$ (customer application)
• output: $y \in Y$ (good/bad after approving credit card)
• unknown pattern to be learned = target function:
• $f : X \to Y$ (ideal credit approval formula)
• data = training examples: $D = \{f(x_1, y_1), (x_2, y_2), \cdots , (x_N, y_N)\}$ (historical records in bank)
• hypothesis = skill with hopefully good performance:
• $g : X \to Y$ (‘learned’ formula to be used)
• target f unknown
• hypothesis g hopefully $\approx$ f, but possibly different from f, (perfection ‘impossible’ when f unknown)
• assume $g \in H = \{h_k\}$
• hypothesis set H:
• can contain good or bad hypotheses
• up to A to pick the ‘best’ one as g

• 最后总结起来就是: 1machine learning : use data to compute hypothesis g that approximates target f

### Jinlai Xu's Profile

My Profile Continue reading

#### Thoughts after 'Value Creation with Data Analytics in the Commonwealth of Virginia'

Published on October 10, 2015

#### Thoughts after 'Research and Education on Ubiquitous Sensing'

Published on October 10, 2015