Machine Learning Foundations
来源:互联网 发布:2017中美人工智能创投 编辑:程序博客网 时间:2024/05/17 03:11
1. when can machines learn?
1.2 Learning to answer yes/no
PLAA takes linear separable D and perceptrons H to get hypothesisg
unknown target function (f: x --> y)
training examplesD: (x1, y1), ···, (xn, yn) ---> learning algorithm A ---> final hypothesis g ≈ f
(hypothesis set H, H = all possible perceptrons)
Perceptron:
A Simple Hypothesis Set: the ‘Perceptron’: called ‘perceptron’ hypothesis historically
Vector Form of Perceptron Hypothesis: h(x) = sign(w^Tx)
Perceptrons in R^2: perceptrons ⇔ linear (binary) classifiers
Perceptron Learning Algorithm (PLA):
start from some w0 (say, 0), and ‘correct’ its mistakes on D
Linear Separability:
linear separable D ⇔ PLA halts (i.e. no more mistakes)⇔ exists perfect wf such that yn = sign(wf^T xn)
More about PLA:
As long as linear separable and correct by mistake
• inner product of wf and wt grows fast; length of wt grows slowly (T <= (R/p)^2)
• PLA ‘lines’ are more and more aligned with wf ⇒ halts
Learning with Noisy Data --> Pocket Algorithm:
modify PLA algorithm (black lines) by keeping best weights in pocket
- Machine Learning Foundations q15
- Machine Learning Foundations
- Machine Learning Foundations
- Machine Learning Foundations
- 機器學習基石 (Machine Learning Foundations)
- Machine Learning Foundations(NTU) 第一次作业
- Machine Learning Foundations(机器学习基石)笔记 第一节
- Machine Learning Foundations(机器学习基石)笔记 第二节
- NTU-Coursera机器学习:机器学习基石 (Machine Learning Foundations)
- Machine Learning Foundations 第1节-第4节
- Machine Learning Foundations 第5节-第8节
- Machine Learning Foundations 第9节-第12节
- Machine Learning Foundations 第13节-第16节
- Foundations : Machine Learning Specialization of University of Washington
- Foundations of Machine Learning 内容整理 —— Introduction
- (机器学习基石)Machine Learning Foundations:Lecture 1
- (机器学习基石)Machine Learning Foundations:Lecture 2
- 【学习总结】Machine Learning Foundations(待补充。。。)
- 欢迎使用CSDN-markdown编辑器
- acm 数塔
- 一个易错的面试题来加深对C++多态的理解
- codeforces 849B. Tell Your World
- 多线程编程核心技术读书笔记(四):Lock的使用
- Machine Learning Foundations
- 完整的RMAN备份示例
- 面试题27. 二叉搜索树与双向链表
- CodeForces
- 增强学习(Reinforcement Learning)
- freemarker中渲染是字符串的替换
- linux文件api
- 自定义条柱图(二)
- Windows文件夹目录快速访问