Introduction:

This article is from deeplearning.ai’s deeplearning course test assignments, which will be translated gradually in the near future.

Translation: Huang Haiguang

Week 1:

Neural Networks and Deep Learning

Week 1 Quiz-Introduction to Deep Learning

1. What does the analogy “AI is the new electricity” refer to? (What’s the parallel to “AI is the new electricity”?)

【 】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.

【 】AI runs on computers and is thus powered by electricity, AI runs on computers and is powered by electricity, but it is making possible things that computers could not do before.

【★】Similar to electricity starting about 100 years ago, AI is a transforming multiple industries. AI is changing many industries.)

Note: Andrew Illustrated illustrated the same idea in the lecture.

2. Which of these are reasons for Deep Learning recently taking off? Check the two options that apply. What are the reasons for the rapid development of deep learning? (Two options)

【★】 We have access to a lot more computational power.

【 】Neural Networks are a brand new field.

【★】 We have access to a lot more data.

【 】Deep learning has significant improvements in important applications such as online advertising, Speech recognition, and image recognition. Deep learning has made significant progress, with many applications in online advertising, speech recognition, and image recognition.

3. Recall this diagram of iterating over different ML ideas. Which of the statements below are true? Check all that apply. Recall an iterative graph of different machine learning ideas. Which (which) of the following statements is true?

【★】Being able to try out ideas quickly allows deep learning engineers to iterate more quickly.

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

【 】It is faster to train on a big dataset than a small dataset. 【 】 Faster to train on a big dataset than a small dataset.

【★】Recent progress in deep learning algorithms has allowed us to train good models faster (even without changing the Using newer deep learning algorithms allows us to train models faster (even if we change CPU/GPU hardware).

Note: A bigger dataset generally requires more time to train on a same  model.

(Note: The same model usually takes more time on larger data sets.)

4. When an experienced deep learning engineer works on a new problem, they can usually use insight from previous problems to train a good model on the first try, without needing to iterate multiple times through different models. True/False? (Is it true that when an experienced deep learning engineer is working on a new problem, they can often use previous experience to train a well-performing model on their first try, rather than iterating through different models several times to select a better one?)

【 】True, True, True

【★】 False

Note: Maybe some experience may help, but nobody can always find the best model or hyperparameters without iterations. Perhaps some previous experience might help, but no one can always find the best model or hyperparameter without iterating multiple times.)

5. Which one of these plots represents a ReLU activation function? (Which of these figures represents ReLU activation?)

Answer:

【 】True, True, True

6. Images for cat recognition is an example of “structured” data, because it is represented as a structured array in a computer. True/False? (Is it true that the image used to identify cats is an example of “structured” data, because it is represented in computers as a structured matrix?)

【 】True, True, True

【★】 False

7. Demographic dataset with statistics on different cities’ population, GDP per capita, Economic growth is an example of “unstructured” data because it contains data coming from different sources. True/False? (Is it true that a demographic data set that counts different city populations, GDP per capita, economic growth is an example of “unstructured” data because it contains data from different sources?)

【 】True, True, True

【★】 False

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

【★】It can be trained as a supervised learning problem.

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

【★】 the input/output is a sequence (e.g., a sequence of words).

【 】RNNs represents the recurrent process of Idea->Code->Experiment->Idea->… RNNs stands for recursive process: Idea -> code -> experiment -> Idea ->…

9. In this diagram which we hand-drew in lecture, what do the horizontal axis (x-axis) and vertical axis (y-axis) represent? (What do the horizontal (x) and vertical (y) axes represent in our hand-drawn picture?)

Answer (the Answer) :

X-axis is the amount of data

Y-axis (vertical axis) is the performance of the algorithm.

Assuming the trends described in the previous question’s figure are accurate (and hoping you got the axis labels) right), which of the following are true? (Assuming that what was described in the previous problem diagram is accurate (and hopefully your axis is labeled correctly), which of the following is true?

【★】Increasing the training set size generally does not hurt an algorithm performance Increasing the size of the training set usually does not affect the performance of the algorithm, which may be of great help.

【★】 Increasing the size of a neural network generally does not hurt an algorithm performance, The performance of the algorithm is significantly improved by increasing the size of neural networks, which may help significantly.

Decreasing the training set size generally does not hurt an algorithm performance, Reducing the size of the training set usually does not affect the performance of the algorithm, which may be of great help.

【 】Decreasing the size of a neural network generally does not hurt an algorithm performance, Significantly, reducing the size of neural networks usually does not affect the performance of algorithms, which may help greatly.

Note: the menu of the official account includes an AI cheat sheet, which is very suitable for learning on the commute.

Highlights from the past2019Machine learning Online Manual Deep Learning online Manual AI Basic Download (Part I) note: To join our wechat group or QQ group, please reply "add group" to join knowledge planet (4500+ user ID:92416895), please reply to knowledge PlanetCopy the code

Like articles, click Looking at the