Day1 机器学习基本概念和梯度下降
机器学习定义
Arthur Samuel:the field of study that gives computers the ability to learn without being explicitly programmed.
Tom Mitchell:A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P, if its performance at tasks in T, as measured by P, improves with experience E.
机器学习常用算法
监督学习(Supervised learning):To predict the result by analyzing some correct relationship between the input and the output.模型如下:
回归问题("regression"):we are trying to predict results within a continuous output.
分类问题("classification"): we are instead trying to predict results in a discrete output.
非监督学习(Unsupervised learning):To approach problems with little or no idea what our results should look like.
代价函数Cost Function
A function used to measure the accuracy of our hypothesis function.
We can use a cost function like
Our target is to find a value of θ0 and θ1 to get the min J
梯度下降算法Gradient Descent Algorithm
repeat until convergence:{
}
α : 学习速率(learning rate).
if α is too small,gradient descent can be slow .
if α is too large,gradient descent can overshoot the minimum.It may fail to converge,or even diverge.
算法中θ0和θ1是同时改变的: