Scribd
Upload
0 views

AI 1

The document outlines the structure and content of mid-term exams for an Artificial Intelligence course, covering topics such as the classification of AI, definitions and relationships between AI, Machine Learning, and Deep Learning, as well as specific algorithms and architectures. It includes questions on logistic vs linear regression, activation functions, and applications of Deep Learning. Additionally, it requires the sketching of a multilayer perceptron and calculation of total neurons in the network.

Uploaded by

Jibel Csay
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0 views

AI 1

The document outlines the structure and content of mid-term exams for an Artificial Intelligence course, covering topics such as the classification of AI, definitions and relationships between AI, Machine Learning, and Deep Learning, as well as specific algorithms and architectures. It includes questions on logistic vs linear regression, activation functions, and applications of Deep Learning. Additionally, it requires the sketching of a multilayer perceptron and calculation of total neurons in the network.

Uploaded by

Jibel Csay
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 5

MID-TERM EXAMS (THEORY)

COURSE: ARTIFICIAL INTELLIGENCE (TRIMESTER 8 AND 9)

1. Classify Artificial intelligence based on its stages and branches?

2. Define Machine Learning and elaborate on the relationship between


Artificial, Machine Learning and Deep Learning?

3. List 5 Machine Learning Algorithms?

4. Elaborate on the differences between logistic and linear regression?

5. Define Deep Learning?

6. List three (3) activation functions used in Deep Learning?

7. List three (3) types of Deep Learning architectures?

18. Provide four (4) applications of Deep Learning?

19. Sketch a multilayer perceptron with three (3) layers where the network
contains 3 input features and the output is a single value. The first layer
contains 5 neurons while the second has 7 neurons.

10. Find the total number of neurons in the multilayer perceptron sketched
above in.
1. Classify Artificial Intelligence based on its stages and branches

Stages of AI:

 Artificial Narrow Intelligence (ANI): Specialized in one task (e.g.,


Siri, chatbots).

 Artificial General Intelligence (AGI): Human-like intelligence, able


to perform any intellectual task.

 Artificial Super Intelligence (ASI): Hypothetical AI surpassing


human intelligence in all fields.

Branches of AI:

 Machine Learning (ML): Systems that learn from data.

 Natural Language Processing (NLP): Understanding and


generating human language.

 Computer Vision: Interpreting and understanding visual input.

 Robotics: AI applied to physical machines.

 Expert Systems: Decision-making systems mimicking human


experts.

 Fuzzy Logic: Reasoning with uncertain or imprecise data.

2. Define Machine Learning and elaborate on the relationship


between Artificial Intelligence, Machine Learning, and Deep
Learning

 Machine Learning: A subset of AI that allows systems to learn from


data and improve over time without being explicitly programmed.

Relationship:

 AI is the broadest concept—creating machines that can simulate


human intelligence.

 ML is a subset of AI—focused on algorithms that allow systems to learn


from data.
 Deep Learning (DL) is a subset of ML—uses neural networks with
many layers to learn from vast amounts of data.

AI ⊇ ML ⊇ DL

3. List 5 Machine Learning Algorithms

1. Linear Regression

2. Logistic Regression

3. Decision Trees

4. Support Vector Machines (SVM)

5. K-Nearest Neighbors (KNN)

4. Elaborate on the differences between Logistic and Linear


Regression

Feature Linear Regression Logistic Regression

Probabilities (0 to 1), classified


Output Type Continuous values
values

Predicting quantities (e.g., Binary classification (e.g., spam


Use Case
price) or not)

Function Linear equation Sigmoid function

Loss
Mean Squared Error Cross Entropy
Function

5. Define Deep Learning

 Deep Learning is a subset of ML that uses neural networks with


multiple layers to model complex patterns in large datasets,
mimicking the human brain's structure.

6. List three (3) activation functions used in Deep Learning

1. ReLU (Rectified Linear Unit)


2. Sigmoid

3. Tanh (Hyperbolic Tangent)

7. List three (3) types of Deep Learning architectures

1. Convolutional Neural Networks (CNNs) – mostly used in image


recognition.

2. Recurrent Neural Networks (RNNs) – used for sequence data like


speech and text.

3. Generative Adversarial Networks (GANs) – used for data


generation like deep fakes.

8. Provide four (4) applications of Deep Learning

1. Facial recognition

2. Autonomous vehicles (self-driving cars)

3. Speech recognition (e.g., Google Assistant)

4. Medical diagnosis (e.g., cancer detection in radiology)

9. Sketch a Multilayer Perceptron

Description:

 Input Layer: 3 neurons (for 3 features)

 Hidden Layer 1: 5 neurons

 Hidden Layer 2: 7 neurons

 Output Layer: 1 neuron

Input: [x1] [x2] [x3]

\ | /

[ H1 - 5 neurons ]

[ H2 - 7 neurons ]
|

[ Output - 1 neuron ]

10. Find the total number of neurons

 Input Layer: 3 neurons

 Hidden Layer 1: 5 neurons

 Hidden Layer 2: 7 neurons

 Output Layer: 1 neuron

Total neurons = 3 + 5 + 7 + 1 = 16 neurons

You might also like