Coursera Machine Learning 第二周 quiz Linear Regression with Multiple Variables 习题答案

1.Suppose m=4 students have taken some class, and the class had a midterm exam and a final exam. You have collected a dataset of their scores on the two exams, which is as follows:

midterm exam(midterm exam)^2 final exam
897921 96
725184 74
948836 87
694761 78

You'd like to use polynomial regression to predict a student's final exam score from their midterm exam score. Concretely, suppose you want to fit a model of the form  hθ(x)=θ0+θ1x1+θ2x2 , where  x1  is the midterm score and x_2 is (midterm score)^2. Further, you plan to use both feature scaling (dividing by the "max-min", or range, of a feature) and mean normalization.

What is the normalized feature  x(4)2 ? (Hint: midterm = 69, final = 78 is training example 4.) Please round off your answer to two decimal places and enter in the text box below.

 平均值为 (7921+5184+8836+4761)/4=6675.5
Max-Min=8836-4761=4075
(4761-6675.5)/4075=-0.46957
保留两位小数为-0.47 

2.You run gradient descent for 15 iterations

with  α=0.3  and compute

J(θ)  after each iteration. You find that the

value of  J(θ)  decreases quickly then levels

off. Based on this, which of the following conclusions seems

most plausible?

Rather than use the current value of  α , it'd be more promising to try a smaller value of  α  (say  α=0.1 ).

α=0.3  is an effective choice of learning rate.

Rather than use the current value of  α , it'd be more promising to try a larger value of  α  (say  α=1.0 ).


下降太快所以a下降速率过大,a越大下降越快,a小下降慢,而本题中,代价函数快速收敛到最小值,代表此时a最合适

Suppose you have  m=14  training examples with  n=3  features (excluding the additional all-ones feature for the intercept term, which you should add). The normal equation is  θ=(XTX)1XTy . For the given values of  m  and  n , what are the dimensions of  θ X , and  y  in this equation?

X  is  14×4 y  is  14×4 θ  is  4×4

X  is  14×3 y  is  14×1 θ  is  3×3

X  is  14×4 y  is  14×1 θ  is  4×1

X  is  14×3 y  is  14×1 θ  is  3×1


此题注意,计算X维度时要加上X0=[1,1,1,1..,1];故答案选C

4

Suppose you have a dataset with  m=50  examples and  n=200000  features for each example. You want to use multivariate linear regression to fit the parameters  θ  to our data. Should you prefer gradient descent or the normal equation?

Gradient descent, since it will always converge to the optimal  θ .

Gradient descent, since  (XTX)1  will be very slow to compute in the normal equation.

The normal equation, since it provides an efficient way to directly find the solution.

The normal equation, since gradient descent might be unable to find the optimal  θ .


训练集有2W,转置的话需要大量的时间,花费太多内存与资源,故选择梯度下降。选择B

Which of the following are reasons for using feature scaling?

It speeds up gradient descent by making it require fewer iterations to get to a good solution.

It speeds up solving for  θ  using the normal equation.

It is necessary to prevent gradient descent from getting stuck in local optima.

It prevents the matrix  XTX  (used in the normal equation) from being non-invertable (singular/degenerate).

迭代次数的减少,加快了正确答案的得出。正规方程对计算只与训练集的大小有关,而与至无关,不能阻止梯度下降局部最优(ps:正规方程没有局部最优),第四个答案,除非可以减少特征变量,否则不能解决此问题所以选A



Coursera机器学习是由斯坦福大学教授Andrew Ng主讲的一门在线课程,旨在向学习者介绍机器学习的基本概念、算法和应用。该课程涵盖了监督学习、无监督学习、深度学习等多个方面,通过理论讲解和实践编程作业,帮助学习者掌握机器学习的基本原理和实践技能。该课程是Coursera平台上最受欢迎的课程之一,也是机器学习领域入门的重要资源之一。 ### 回答2: Coursera机器学习是由斯坦福大学Andrew Ng教授设计并教授的在线课程。这门课程被认为是公认的机器学习入门教材之一,也是Coursera平台最受欢迎的课程之一。 这门课程涵盖了机器学习领域中最基础的知识和技术,包括监督学习、无监督学习以及神经网络等。学生可以通过该课程了解到如何采集和处理数据、如何训练模型、如何评估模型的性能等。此外,课程还涉及到机器学习中一些实用的技术,如正则化、梯度下降等。 该课程受到了全球范围内的认可和青睐,许多学生、工程师、数据科学家等都受益于该课程。由于该课程的知识点全面、深入浅出、容易理解和学习,在业内和学术界都广受赞誉,拥有较高的知名度和价值。 总之,Coursera机器学习是一门非常好的课程,对于那些对机器学习感兴趣的人来说,它是一个不可错过的机会。课程教材内容丰富、难度适中,且教学相对轻松愉悦,难怪在学习资源上产生了广泛的影响。
评论 4
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值