Machine Learning Course 1 and 2

What is the Machine Learning？

​ Experience E ， Task T， Performance measure P

​ If its performance on T as measured by P , improves with experience E。

​ 经验 E, 任务 T, 性能度量 P。

​ 由 P 定义的关于 T 的性能随着 E经验而提高。

Machine Learning Algorithm

• Supervised learning （监督学习， 告诉计算机什么是正确的， 什么是错误的）
• Unsupervised learning （由计算机自己学习如何正确）
• Reinforcement learning （增强学习）
• Recommender Systems （推荐系统）

Supervised Learning

​ “right answers” Given

​ Regression： Predict continuous valued output

​ 回归问题： 预测一个连续的输出值。

​ Classification： Discrete valued output（0 or 1）

​ 分类问题： 预测离散的输出值。

Unsupervised Learning

​ cluster 聚类.

​ 聚类问题。

​ Clustering algorithm

Linear regression with one variable

Model representation

​ fit a model 拟合模型

···

​ $h_{\theta} (x) = {\theta}_0 + {\theta}_1 x$

​ Linear regression with one variable or univariate linear regression

​ 单变量线性回归。

​ $h_{\theta} (x) = {\theta}_0 + {\theta}_1 x$

​ $\theta_{i^`}$ : Parameters 模型参数。

Cost function

​ squared error cost function $J(\theta_0, \theta_1) =\frac{1}{2m} \sum_{{i=1}}^{m} (h_{\theta}(x^{(i)} - y^{(i)})^2$ $minimaze_{\theta_{0}, \theta_{1}} J (\theta_{0}, \theta_{1})$

Gradient descent

​ 梯度下降算法， 找到 $Min\{ J(\theta_0, \theta_1) \}$

$$repeat\quad until\quad convergence :\\ \theta_{j} := \theta_{j} - \alpha\frac{\partial}{\partial\theta_{j}}J(\theta_{0}, \theta_{1})\\ $$

---

​ Correct : Simultaneous update

$$temp0 := \theta_{0} - \alpha\frac{\partial}{\partial\theta_{0}}J(\theta_{0},\theta_{1})\\ temp1 := \theta_{1} - \alpha\frac{\partial}{\partial\theta_{1}}J(\theta_{0},\theta_{1})\\ \theta_{0} := temp0; \\ \theta_{1} := temp1$$

​Gradient descent algorithm

$$repeat \quad until \quad convergence:\\ \theta_{0} := \theta_{0} - \alpha\frac{1}{m}\sum_{i=1}^{m} (h_\theta(x^{(i)} - y^{(i)}))\\ \theta_{1} := \theta_{1} - \alpha\frac{1}{m}\sum_{i=1}^{m} (h_\theta(x^{(i)} - y^{(i)}))*x^{(i)}$$

​ “Batch” : Each step of gradient descent uses all the training examples.

​ 每一次梯度下降，都使用所有的训练集合。