Question 1: Choose the correct answer:

Neural Network
QUESTION NO: 1
Which of the following does not belong to automatic hyperparameter optimization
algorithm?
A. Grid search
B. Random gradient descent
C. Random search
D. Model-based hyperparameter optimization
Answer: B
QUESTION NO: 147
Which description is wrong about the hyperparameter?
A. Hyperparameters are parameters that set values before the algorithm begins learning.
B. Most machine learning algorithms have hyperparameters.
C. Hyperparameters cannot be modified.
D. The value of the hyperparameter is not learned by the algorithm itself.
Answer: C
QUESTION NO: 9
Which of the following does not belong to long-term memory LSTM (Long Short-Term
Memory) network architecture?
A. Memory door
B. Forget the door
C. Input door
D. Output door
Answer: A
QUESTION NO: 14
What are the commonly used activation functions? (Multiple Choice)
A. sigmoid
B. tanh
C. relu
D. danish
Answer: ABC
QUESTION NO: 27
Deep learning is different from machine learning and there are no unsupervised
algorithms.
A. True
B. False
Answer: B
1 Dr. Fatma M. Talaat
2 Dr. Fatma M. Talaat
QUESTION NO: 31
What is wrong description of backpropagation?
A. The learning process of the backpropagation algorithm consists of a forward
propagation
process and a back-propagation process.
B. The backpropagation algorithm is a learning algorithm suitable for multi-layer neural
networks, which is based on the gradient descent method.
C. The backpropagation phase sends training inputs to the network to obtain an stimuli
response.
D. The backpropagation algorithm is mainly repeated by two loops (excitation
propagation,
weight update) until the response of the network to the input reaches the predetermined
target range.
Answer: C
QUESTION NO: 38
Deep learning makes it easy to derive simple mathematical functions from a large amount
of high-dimensional data to describe complex decision interfaces.
A. True
B. False
Answer: B
QUESTION NO: 39
What are the mainstream deep learning open source tools? (Multiple choice)
A. Theano
B. Caffe
C. TensorFlow
D. Scikit-learn
E. Torch
Answer: A,B,C,E
QUESTION NO: 124
Which of the following is not part of the deep learning development framework?
A. CNTK
B. Keras
C. BAFA
D. MXNet
Answer: C
QUESTION NO: 45
What are the commonly used loss functions? (Multiple choice)
A. Mean variance
B. Sigmoid cross entropy
C. Softmax cross entropy
3 Dr. Fatma M. Talaat
D. Sparse cross entropy
E. Weighted Sigmoid Cross Entropy
Answer: ABCDE
QUESTION NO: 48
Which of the following is not included in the recurrent neural network usage scenario?
A. Machine translation
B. Speech recognition
C. Image style migration
D. Text generation
Answer: C
QUESTION NO: 56
Cyclic neural networks are more suitable for dealing with image recognition problems.
A. True
B. False
Answer: B
QUESTION NO: 166
Convolutional neural networks are more suitable for image recognition problems than
cyclic neural networks.
A. True
B. False
Answer: A
QUESTION NO: 167
What model is not a cyclic neural network?
A. RNN
B. LSTM
C. GBDT
D. GRU
Answer: C
What is not the optimization method in deep learning?
A. Random gradient descent
B. Back propagation algorithm
C. Principal component analysis
D. Momentum
Answer: C
QUESTION NO: 142
What does not belong to supervised learning?
A. Logistic regression
B. Support vector machine
4 Dr. Fatma M. Talaat
C. Decision tree
D. Principal component analysis
Answer: D
QUESTION NO: 61
In a neural network, knowing the weight and deviations of each neuron is the most
important step. If you know the exact weights and deviations of neurons in some way,
you can approximate any function.
What is the best way to achieve this?
A. Random assignment, pray that they are correct
B. Search for a combination of weight and deviation until the best value is obtained
C. Assign an initial value to iteratively update weight by checking the difference
between the best value and the initial.
D. The above is not correct
Answer: C

5 Dr. Fatma M. Talaat

QUESTION NO: 64
Machine Learning Service is a data mining and analytics platform service that helps users
quickly discover data patterns and build predictive models through machine learning
techniques and deploy them as predictive analytics solutions.
A. True
B. False
Answer: A
QUESTION NO: 143
Which of the following is true about unsupervised learning?
A. Unsupervised algorithm only processes "features" and does not process "tags"
B. Dimensionality reduction algorithm is not unsupervised learning
C. K-means algorithm and SVM algorithm belong to unsupervised learning
D. None of the above
Answer: A
QUESTION NO: 66
Training error will reduce the accuracy of the model and produce under-fitting. How to
improve the model fit? (Multiple choice)
A. Increase the amount of data
B. Feature Engineering
C. Reduce regularization parameters
D. Add features
Answer: ABD
QUESTION NO: 72
Reducing the gap between the training error and the test error will result in over-fitting.
How
to prevent over-fitting? (Multiple Choice)
A. Cross validation
B. Integration method
C. Increase regularization
D. Feature Engineering
Answer: ABCD
QUESTION NO: 75
Which of the following can optimization problem be classified according to the
constraints?
(Multiple choice)
A. equality constraints
B. Inequality constraints
C. Unconstrained conditions
D. Semi-constraint conditions
Answer: ABC
6 Dr. Fatma M. Talaat
QUESTION NO: 123
What are the regularizations in deep learning? (Multiple Choice)
A. L1 norm, L2 norm
B. Data set enhancement
C. Integration method
D. Dropout
Answer: ABCD
QUESTION NO: 77
Deep learning algorithms can be divided into supervised learning and unsupervised
learning.
A. True
B. False
Answer: A

7 Dr. Fatma M. Talaat

QUESTION NO: 81
What is the incorrect relationship between neural networks and deep learning?
A. The concept of deep learning stems from the study of artificial neural networks.
B. A neural network algorithm with multiple hidden layers is a deep learning algorithm.
C. Single layer neural networks are also a type of deep learning.
D. Convolutional neural networks are a type of deep learning.
Answer: C
QUESTION NO: 84
Which of the following can improve the computational efficiency of the neural network
model?
(Multiple choice)
A. TPU
B. GPU
C. FPGA
D. Large-scale distributed cluster
Answer: ABCD
QUESTION NO: 85
Which is not a deep learning algorithm?
A. Self-encoder
B. Convolutional neural networks
C. Recurrent neural networks
D. Support vector machine
Answer: D
QUESTION NO: 126
Self-encoder is an unsupervised learning algorithm.
A. True
B. False
Answer: A
QUESTION NO: 96
Which of the following descriptions of the depth feedforward network is correct?
A. Deep feedforward network is a kind of neural network.
B. The deep feedforward network has only one hidden layer.
C. The unit on the hidden layer of the deep feedforward network will have countless.
D. Deep feedforward network is used to deal with linear problems.
Answer: A
QUESTION NO: 102
Which of the following descriptions about the Recurrent Neural Network (RNN) is
correct?
A. Can be used to process sequence data.
8 Dr. Fatma M. Talaat
B. Cannot process variable length sequence data.
C. Unlike convolutional neural networks, parameters of RNN cannot be shared.
D. The units above the hidden layer are not associated with each other.
Answer: A
QUESTION NO: 174
Which of the following neural network structures will share weights? (Multiple choice)
A. Convolutional neural network
B. Recurrent neural network
C. Fully connected neural network
D. All of the above
Answer: AB
QUESTION NO: 107
Which of the following options is not a reason for traditional machine learning
algorithms to promote the development of deep learning?
A. Dimensional disaster
B. Local invariance and smooth regularization
C. Manifold learning
D. Feature Engineering
Answer: D
QUESTION NO: 108
Which of the following description of the validation set is wrong?
A. The verification set can coincide with the test set.
B. The test set can coincide with the training set.
C. The subset used to pick hyperparameters is called a validation set.
D. Typically 80% of the training data is used for training and 20% is used for
verification.
Answer: A
QUESTION NO: 119
Convolutional neural networks are more suitable for dealing with speech recognition
problems.
A. True
B. False
Answer: B
QUESTION NO: 135
There are many commercial applications for machine learning services. What are the
main business scenarios covered? (Multiple Choice)
A. Financial product recommendation
B. Predictive maintenance
C. Telecom customer retention
9 Dr. Fatma M. Talaat
D. Retailer grouping
Answer: ABCD
QUESTION NO: 162
What of the following does belong to convolutional neural network (CNN)? (Multiple
Choice)
A. VGGNet
B. ResNet
C. AlexNet
D. GoogleNet
Answer: ABCD

10 Dr. Fatma M. Talaat

QUESTION NO: 164
Which of the following does the convolutional neural network structure not include?
A. Convolutional layer
B. Pooling layer
C. Loop layer
D. Fully connected layer
Answer: C
QUESTION NO: 171
Deep learning is a branch of machine learning.
A. True
B. False
Answer: A

11 Dr. Fatma M. Talaat

TensorFlow
QUESTION NO: 2
Which method is not supported in TensorFlow to define variables?
A. Random number
B. Constant
C. Calculated from the initial values of other variables
D. null
Answer: D

QUESTION NO: 4
What are the algorithms supported by TensorFlow? (Multiple Choice)
A. CNN
B. ZNN
C. RNN
D. HUT
Answer: AC
QUESTION NO: 12
Which of the following is not a module in the TensorFlow library?
A. tf.nn
B. tf. layers
C. tf. boost
D. tf.contrib
Answer: C
QUESTION NO: 23
Which of the following options is not the session mode used by TensorFlow?
A. Explicitly call the session to generate function
B. Explicitly call the session to close function
C. Through the Python context manager
D. Multiple POST queries
Answer: D
QUESTION NO: 33
What are the implementation modes of TensorFlow? (Multiple Choice)
A. Stand-alone mode
B. Distributed mode
C. Reverse mode
D. Forward mode
Answer: AB

12 Dr. Fatma M. Talaat

QUESTION NO: 34
Which of the following is not a way for the TensorFlow program to read data?
A. Preload data
B. Feeding data
C. Read from the file
D. Write a file format reader
Answer: D
QUESTION NO: 35
Which of the following options is not a tensor attribute?
A. name (Logo)
B. shape (Dimensionality)
C. type (Type)
D. node (Node)
Answer: D
QUESTION NO: 43
Which of the following are included in the application scenarios of TensorFlow?
(Multiple choice)
A. Speech recognition
B. Face recognition
C. Image style changes
D. Autopilot
Answer: ABCD
QUESTION NO: 60
TensorFlow is the second generation of artificial intelligence learning system developed
by
Google based on ( ).
A. DistBelief
B. PaleyFunction
C. ConvexOne
D. Infinity
Answer: A
QUESTION NO: 173
TensorFlow supports multi-TPU cluster computing.
A. True
B. False
Answer: A

13 Dr. Fatma M. Talaat

QUESTION NO: 62
Which of the following options belong to TensorFlow concept? (Multiple Choice)
A. Tensor
B. Variables
C. Placeholder
D. Operation
E. Conversation
Answer: ABCDE
QUESTION NO: 120
Which of the following is the computing model of TensorFlow?
A. Calculation chart
B. Tensor
C. Conversation
D. Variable
Answer: A
QUESTION NO: 169
In TensorFlow, data is represented in the form of tensors and calculation charts.
A. True
B. False
Answer: B
QUESTION NO: 46
TensorFlow Operations and Computation Graph are not - run in the Session.
A. True
B. False
Answer: B

QUESTION NO: 144

Which of the following options is not the TensorFlow build process?
A. Building a calculation chart
B. Input tensor
C. Generate a session
D. Update weights
Answer: D
QUESTION NO: 82
There are a lot of data generated during the training of the neural network. What
mechanism does TensorFlow use to avoid excessive input data?
A. Client
B. feed
14 Dr. Fatma M. Talaat
C. placeholder
D. fetch
Answer: C

QUESTION NO: 76
Which of the following environments does not support to install the TensorFlow?
A. Linux
B. Mac OS
C. Docker
D. OpenStack
Answer: D
QUESTION NO: 91
The TensorFlow framework does not support Windows systems.
A. True
B. False
Answer: B
QUESTION NO: 170
Which of the following are the characteristics of TensorFlow? (Multiple choice)
A. Open source
B. Convenience
C. Mature
D. Flexible
Answer: ABCD
QUESTION NO: 83
Which of the following data types does TensorFlow not support?
A. int8
B. float32
C. double64
D. bool
Answer: C

QUESTION NO: 89
TensorFlow only supports Python and R language
A. True
B. False
Answer: B
QUESTION NO: 90
What are the core concepts in TensorFlow? (Multiple Choice)

15 Dr. Fatma M. Talaat

A. Gridding
B. Calculation Chart
C. Dot Product
D. Tensor
Answer: BD
QUESTION NO: 97
Which of the following files is the graph structure of TensorFlow saved in?
A. checkpoint file
B. index file
C. .cpkt file
D. .meta file
Answer: D
QUESTION NO: 141
The number or matrix is stored in the tensor.
A. True
B. False
Answer: B
QUESTION NO: 158
TensorFlow is Google's first generation of proprietary machine learning systems.
A. True
B. False
Answer: B

16 Dr. Fatma M. Talaat

Reinforcement Learning
Q1. The main functions of intelligent agents are:
a) Sensors and Actuators
b) Classification, Regression, Clustering
c) Perception and Actions
d) Actuators and Effectors

Q2. ___________ are agents have the ability to gradually improve and
become more knowledgeable about an environment over time through an
additional learning element.
a) Utility-based agents
b) Learning agents
c) Goal-based agents
d) Model-based agents

Q3. Which instruments are used for perceiving and acting upon the
environment?
a) Sensors and Actuators
b) Sensors
c) Perceiver
d) None of the mentioned

Q4. What is the rule of simple reflex agent?

a) Simple-action rule
b) Condition-action rule
c) Simple & Condition-action rule
d) None of the mentioned

17 Dr. Fatma M. Talaat

Q5. What are the composition for agents in artificial intelligence?
a) Program
b) Architecture
c) Both Program & Architecture
d) None of the mentioned

Q6. In which agent does the problem generator is present?

a) Learning agent
b) Observing agent
c) Reflex agent
d) None of the mentioned

Q7. Which is used to improve the agent’s performance?

a) Perceiving
b) Learning
c) Observing
d) None of the mentioned

Q8. Which agent deals with happy and unhappy states

a) Simple reflex agent
b) Model based agent
c) Learning agent
d) Utility based agent

Q9. Which action sequences are used to achieve the agent’s goal?
a) Search
b) Plan

18 Dr. Fatma M. Talaat

 c) Retrieve
d) Both Search & Plan

Q10. Which element in the agent are used for selecting external actions?
a) Perceive
b) Performance
c) Learning
d) Actuator

Q11. The action of the Simple reflex agent completely depends upon
__________.
a) Perception history
b) Current perception
c) Learning theory
d) Utility functions

Q12. What among the following is/are the example of the intelligent
agent/agents?
a) Human
b) Robot
c) Autonomous Spacecraft
d) All of the mentioned

Q13. Wumpus World is a classic problem, best example of _______.

a) Knowledge based Game
b) Reasoning with Knowledge
c) Single player Game
d) Two player Game

19 Dr. Fatma M. Talaat

Q18. Uncertainty arises in the wumpus world because the agent’s sensors
give only ___________.
a) Full & Global information
b) Partial & Global Information
c) Partial & local Information
d) Full & local information

Q19. _________ deals with agents that must sense & act upon their
environment. This combines classical AI and machine learning techniques.
a) Semi supervised Learning
b) ANN
c) Deep Learning
d) Reinforcement Learning

Q20. ______________ is an immediate return sent back from the

environment to evaluate the last action by the agent.
a) State
b) Reward
c) Action
d) None of the above

Q21. ___________________ is the strategy that the agent employs to

determine next action based on the current state.
a) Policy (π)
b) Value (V)
c) Q-value (Q)
d) Action (a)

20 Dr. Fatma M. Talaat

Q22. There are majorly three approaches to implement RL:
a) Model-based, Environment-based, Action-based
b) Value-based, Reward-based, Action-based
c) Value-based, Policy-based, Model-based
d) Policy-based, Model-based, Action-based

Q22. Reinforcement learning is ________________.

a) Unsupervised learning
b) Supervised learning
c) Award based learning
d) None

Q23. Which of the following is an application of reinforcement learning?

a) Topic modeling
b) Recommendation system
c) Pattern recognition
d) Image classification

Q24. There are two types of policy-based RL methods:

a) Deterministic and value-based
b) Deterministic and Stochastic
c) Non-Deterministic and Stochastic
d) static and dynamic

4) ……………… is the choice of action for each state.

a) Reward
b) Value
c) Policy
d) Utility
21 Dr. Fatma M. Talaat
Question 2: Choose the correct answer (True or False):
Q1. One applications of intelligent agents is Information search, retrieval,
and navigation.
a) True
b) False

Q2. AI assistants, like Alexa and Siri, are examples of intelligent agents as
they use sensors to perceive a request made by the user and they
automatically collect data from the internet with some user's help.
a) True
b) False

Q3. Autonomous vehicles could not be considered intelligent agents.

a) True
b) False

Q4. Repetitive office activities cannot be considered as intelligent agents as

there is no any creativity in them.
a) True
b) False

22 Dr. Fatma M. Talaat

Q5. Anything that recognizes the environment through sensors and acts
upon the environment initiated through actuators is called AGENTS.
a) True
b) False

Q6. Agents that extract any specifiable information, such as included

keywords or publication date is called searching agents.
a) True
b) False

Q7. Intelligent agents can be used for applications in data mining, data
analytics and customer service and support (CSS).
a) True
b) False

Q8. In medical diagnosis, the patient is considered as an agent.

a) True
b) False

Q9. Wumpus World is a classic problem, best example of Knowledge based

Game.
a) True
b) False

23 Dr. Fatma M. Talaat