吴恩达深度学习测验题：deeplearning.ai-week1-quiz

Introduction to Deep Learning

答案在中文题目处展示，仅供参考，不一定是正确答案。如有错误，欢迎评论区指出。

英文原题

1. What does the analogy “AI is the new electricity” refer to?

·AI runs on computers and is thus powered by electricity, but it is letting computers do things not possible before.

·AI is powering personal devices in our homes and offices, similar to electricity.

·Through the “smart grid”, AI is delivering a new wave of electricity.

·Similar to electricity starting about 100 years ago, AI is transforming multiple industries.

2. Which of the following are reasons that didn’t allow Deep Learning to be developed during the '80s?

·People were afraid of a machine rebellion

·The theoretical tools didn’t exist during the 80’s

·Interesting applications such as image recognition require large amounts of data that were not available.

·Limited computational power.

3. Recall this diagram of iterating over different ML ideas. Which of the statements below are true? (Check all that
   apply.)

·It is faster to train on a big dataset than a small dataset.

·Being able to tny out ideas quicdly llos deep learming engineers to iterate more quicky.

·Faster computation can help speed up how long a team takes to iterate to a good idea.

·Recent progress in deep learning algorithms has allowed us to train good models faster (even without changing the CPU/GPU hardware).

4. When bilding a neural network to pedict housing price from faures like size, the number of bedrooms, zip coed and wealth,it is necessary to come up with other features in between input and output like family size and school quality. True/False?

·False

·True

5. Which one of these plots reesnts a ReLU ativation functio?
   ·Figure 1:
   
   Figure 3:
   
   Figure 2:
   
   Figure 4:
6. Features of animals, such as weight, height, and color, are used for classification between cats, dogs, or others.This is an example of “structured” data, because they are represented as arrays in a computer. True/False?

·True
Yes. The data can be represented by columns of data. This is an example of structured data.unlike images of the animal.

·False
No. The data can be represented by columns of data. This is an example of structured data.unlike images of the animal.

7. A dataset is composed of age and weight data for several people. This dataset is an example of “structured” data
   because it is represented as an array in a computer.True/False?

·False

·True

8. Why is an RNN (Recurrent Neural Network) used for machine translation, say translating English to French? (Check all that apply.)

·RNNs represent the recurrent process of ldea->Code->Experiment->ldea->……

·It can be trained as a supervised learning problem.

·It is applicable when the input/output is a sequence (eg- a sequence of words).

·It is strictly more powerful than a Convolutional Neural Network (CNN).

9. Suppose the information given in the diagram is accurate. We can deduce that when using large training sets, for a
   model to keep improving as the amount of data for training grows, the size of the neural network must grow. True/False?

·False

·True

10. Assuming the trends dscibed in the figure are accurate Which of the flowing satemets are true? Choose all that apply.

·Increasing the training set size of a traditional learning algorithm always improves its performance.

·Increasing the size of a neural network generally does not hurt an algorithm’s performance, and it may help signifcantly.

·Decreasing the training set size generally does not hurt an algorithm’s performance. and it may help significantly.

·Increasing the training set size of a traditional learning algorithm stops helping to improve the performance after a certain size.

中文题目

1. 和“AI是新电力”相类似的说法是什么？

()AI在计算机上运行，​​并由电力驱动，但是它正在让以前的计算机不能做的事情变为可能

()AI为我们的家庭和办公室的个人设备供电，类似于电力。

()通过“智能电网”，AI提供新的电能。

(√)就像100年前产生电能一样，AI正在改变很多的行业。

2. 下面哪些是80年代深度学习没有发展起来的原因？ (两个选项)

（） 人们害怕机器造反。

（）理论根据在80年还不存在。

（√） 有趣的应用程序，如图像识别，需要大量数据，但还无法获取足够的数据。

（）计算能力有限。

3. 回想一下关于不同的机器学习思想的迭代图。下面哪(个/些)陈述是正确的？

（）在大数据集上训练上的时间要快于小数据集。

（√）能够让深度学习工程师快速地实现自己的想法。

（√）在更好更快的计算机上能够帮助一个团队减少迭代(训练)的时间。

（√） 使用更新的深度学习算法可以使我们能够更快地训练好模型（即使更换CPU / GPU硬件）。

4. 当建立一个神经网络来预测房价的特征，如大小，卧室数量，邮编和财富，有必要在投入和产出之间提出其他特征，如家庭规模和学校质量。对/错?

()正确
(√)错误

5. 这些图中的哪一个表示ReLU激活功能？
   ()Figure 1:
   
   (√)Figure 3:
   
   ()Figure 2:
   
   ()Figure 4:
   

6. 动物的特征，如体重、身高和颜色，被用来区分猫、狗或其他动物。这是“结构化”数据的一个例子，因为它们在计算机中被表示为数组。对/错?

(√)对
是的，数据可以用数据列表示。这是结构化数据的一个例子。不像动物的图像。

()错
数据可以用数据列表示。这是结构化数据的一个例子。不像动物的图像。

7. 一个数据集由几个人的年龄和体重数据组成。这个数据集是“结构化”数据的一个例子，因为它在计算机中以数组的形式表示。对/错?

(√)对

()错

8. 为什么在上RNN（循环神经网络）可以应用机器翻译将英语翻译成法语？

()RNNs代表递归过程：想法->编码->实验->想法->…

(√) 因为它可以被用做监督学习。

(√)它比较适合用于当输入/输出是一个序列的时候（例如：一个单词序列）

()严格意义上它比卷积神经网络（CNN）效果更好。

9. 假设图中给出的信息是准确的。我们可以推断，当使用大的训练集时，对于一个模型来说，随着用于训练的数据量的增长而不断提高，神经网络的大小必须增长。对/错?

(√)对
() 错

10. 假设图中描述的趋势是准确的。下列哪个陈述是正确的?选择所有适合的。

() 增加传统学习算法的训练集大小总是能提高算法的性能。

(√) 增加神经网络的大小通常不会损害算法的性能。这可能会有很大的帮助。

()减少训练集的大小通常不会损害算法的性能，而且可能会有显著的帮助。

(√)对于传统的学习算法，增加训练集的大小到一定的大小后就不再有助于提高性能。