机器学习基石上 (Machine Learning Foundations)—Mathematical Foundations
Hsuan-Tien Lin, 林轩田,副教授 (Associate Professor),资讯工程学系 (Computer Science and Information Engineering)
Feasibility of Learning
Learning is Impossible?
Two Controversial Answers 多种合理的方式得到不同的答案
no-free-lunch problems
Probability to the Rescue
Inferring Something Unknown → sample
样本(in-sample)的概率与整体(out-of-sample)的概率大概是接近的
Hoeffding’s Inequality
Connection to Learning
抓弹珠类比学习
- 抓的一把弹珠是已知数据
- 橙色的代表错误
- 抽样测试,测试集上的正确率
增加部件
公式表述
Verification
- Verification of One h
- The Verification Flow
- Verification of One h
Connection to Real Learning
BAD sample: Ein E i n and Eout E o u t far away (can get worse when involving choice)
BAD Data for One h: Ein(h) E i n ( h ) and Eout(h) E o u t ( h ) far away
- 不好的几率很小
- 不好的几率很小
BAD Data for Many h
- for
M
M
hypotheses, bound of
- for
M
M
hypotheses, bound of
The Statistical Learning Flow
learning possible if |H| | H | finite and Ein(g) E i n ( g ) small