Chapter 2: Infrastructure and Tools For Ai

Question 1

What does an operator take care of?

Answer 1

Conditions and everything that affects them

It consists of actions, preconditions, and changes resulting from taking actions.

Question 2

What is a rational agent?

Answer 2

An entity that takes actions to achieve certain goals

The text discusses making rational agents more intelligent.

Question 3

What are common techniques to impart intelligence to an agent?

Answer 3

• Machine learning
• Stored knowledge
• Rules

The focus is on machine learning in the text.

Question 4

What is the role of labeled data in machine learning?

Answer 4

To solve a given problem by allowing machines to learn patterns

Machines extract patterns and relationships from labeled data.

Question 5

What are the steps that an intelligent agent takes when using machine learning?

Answer 5

• Sensor perceives input
• Sends to feature extraction block
• Trained inference engine performs prediction
• Inference engine sends decision to actuator

This process results in taking required action in the real world.

Question 6

Name some applications of machine learning.

Answer 6

• Image recognition
• Robotics
• Speech recognition
• Predicting stock market behavior

Machine learning is widely used in various fields.

Question 7

What are the two types of models in AI?

Answer 7

• Analytical models
• Learned models

The text discusses the transition from analytical to learned models.

Question 8

What characterizes analytical models?

Answer 8

Derived using mathematical formulation and human judgment

They are often simplistic and inaccurate.

Question 9

What defines learned models?

Answer 9

Obtained through training, analyzing many examples of inputs and outputs

These models are complex and accurate.

Question 10

What is a significant advantage of machine learning over analytical models?

Answer 10

No need to derive the underlying mathematical formula

Machines derive the formula based on data.

Question 11

What is the first command to enter the Python command prompt?

Answer 11

$ python3

This command starts the Python interpreter.

Question 12

What did Alan Turing ask in 1950?

Answer 12

Can machines think?

This question remains relevant in AI discussions today.

Question 13

What does ML stand for?

Answer 13

Machine Learning

It is a key area of focus in applied AI.

Question 14

What is the difference between strong AI and weak AI?

Answer 14

• Strong AI: Refers to AGI (Artificial General Intelligence)
• Weak AI: Refers to ANI (Artificial Narrow Intelligence)

Weak AI is what is generally meant in most product contexts.

Question 15

True or False: AGI exists today.

Answer 15

False

AGI does not exist; current AI is narrow in application.

Question 16

What is the main characteristic of ANI?

Answer 16

Narrow application of AI that is good at specific tasks

Its expertise does not extend to other areas.

Question 17

What is a rules-based engine?

Answer 17

A system that replicates human decision-making without learning

It is considered AI but not ML.

Question 18

What is the main challenge in AI adoption mentioned in the text?

Answer 18

Finding clear information about what makes a product AI

Marketing often complicates the understanding of AI.

Question 19

What does AGI stand for?

Answer 19

Artificial General Intelligence

AGI refers to a type of AI that can understand, learn, and apply intelligence across a wide range of tasks, similar to human cognitive abilities.

Question 20

True or False: Sentient AI currently exists.

Answer 20

False

Sentient AI does not exist, and the statement emphasizes the importance of ethical AI applications.

Question 21

What are the consequences of unethical applications of AI?

Answer 21

They cause minor inconveniences and major upsets

Unethical applications of AI can lead to significant harm in various contexts.

Question 22

What are the two essential components of Machine Learning (ML)?

Answer 22

Models and training data

The models are used to learn from historical data points.

Question 23

What is the role of data scientists and ML engineers in ML?

Answer 23

They choose, build, tune, and maintain models for optimized performance

Their work is crucial for ensuring that ML models function effectively.

Question 24

What are the four major learning categories in ML?

Answer 24

• Supervised learning
• Unsupervised learning
• Semi-supervised learning
• Reinforcement learning

Each learning type has specific models and algorithms associated with it.

Question 25

What differentiates Deep Learning (DL) from Machine Learning (ML)?

Answer 25

DL is based on neural network algorithms, while ML encompasses other algorithms ## Footnote DL is considered a subset of ML.

Question 26

What does the term 'feature selection' refer to in the context of DL?

Answer 26

The neural networks learn to pick up patterns or features from the data autonomously ## Footnote This contrasts with traditional ML where engineers select features.

Question 27

What is supervised learning?

Answer 27

Humans label the data, and machines attempt to label current or future data points ## Footnote This method involves continuous training to improve model accuracy.

Question 28

Name one application of supervised learning models.

Answer 28

* Classification models (e.g., spam filters) * Regression models (e.g., predicting trends) ## Footnote These applications require continuous training and updating.

Question 29

What is the Naive Bayes algorithm used for?

Answer 29

Classification problems ## Footnote It considers each feature in the dataset as an independent variable.

Question 30

What does SVM stand for?

Answer 30

Support Vector Machine ## Footnote SVM is used for classification by splitting datasets into two classes.

Question 31

What is the purpose of linear regression?

Answer 31

To predict future data points based on one or more variables ## Footnote It finds the best line that fits the data.

Question 32

What does logistic regression predict?

Answer 32

A future binary categorical state ## Footnote Examples include predicting loan defaults.

Question 33

How does a decision tree model work?

Answer 33

It uses nodes and branches to learn from past data to predict future values ## Footnote Decision trees are popular due to their interpretability.

Question 34

What is a random forest?

Answer 34

An ensemble of decision trees used for both categorical and numerical predictions ## Footnote It averages or votes for predictions based on multiple trees.

Question 35

What characterizes unsupervised learning?

Answer 35

The data is unlabeled, and machines find patterns independently ## Footnote This method is often used when the correct answers are unknown.

Question 36

Name one application of unsupervised learning.

Answer 36

* Clustering * Dimensionality reduction ## Footnote Clustering identifies groups in data, while dimensionality reduction simplifies datasets.

Question 37

What does PCA stand for, and what is its purpose?

Answer 37

Principal Component Analysis; reduces dimensions without losing information ## Footnote PCA is useful for analyzing large datasets.

Question 38

What is semi-supervised learning?

Answer 38

A model trained with both labeled and unlabeled data ## Footnote It helps guide the model towards finding patterns when labeled data is scarce.

Question 39

What is the primary challenge of overfitting in ML?

Answer 39

When a model fits too well to a specific dataset and performs poorly on new data ## Footnote This is a common issue in data science.

Question 40

What is semi-supervised learning?

Answer 40

A learning method that uses a small amount of labeled data alongside a larger amount of unlabeled data to improve model performance ## Footnote The process involves predicting on unlabeled data and checking accuracy against the labeled data.

Question 41

How does reinforcement learning operate?

Answer 41

It learns through trial and error, adapting its approach based on past behavior and optimizing for rewards ## Footnote Commonly used in robotics to help machines adjust to real-world parameters.

Question 42

What are the main components of operationalizing AI/ML?

Answer 42

* Choosing models and use cases * Managing production * Updating models * Keeping data fresh and clean * Organizing experiments * Validating and testing ## Footnote This process is complex and often a challenge for companies outside the tech industry.

Question 43

What is the purpose of an ETL pipeline?

Answer 43

To make layers of data and metadata available for AI/ML models to ingest and offer insights from ## Footnote ETL stands for Extract, Transform, Load.

Question 44

What are the four major components of the continuous maintenance process in AI?

Answer 44

* Testing/validating code and components * Continuous code updates * Continuous learning from new data * Monitoring model performance ## Footnote This ensures that models do not become stale and maintain effective performance.

Question 45

Why is proper data storage critical for AI/ML products?

Answer 45

It impacts the performance of models and the overall product ## Footnote Poor data storage can lead to significant operational issues.

Question 46

What are the differences between a data lake, database, and data warehouse?

Answer 46

* Data Lake: Stores raw data in its native format * Database: Structured data for easy access and querying * Data Warehouse: Centralizes structured data for analysis and insights ## Footnote Each serves different purposes based on organizational needs.

Question 47

What is the benefit of using a data warehouse for AI/ML?

Answer 47

It allows for quick leverage of insights and trends across various business units ## Footnote Essential for organizations looking to implement AI/ML across multiple functions.

Question 48

True or False: A relational database is ideal for combining various datasets for advanced analytics.

Answer 48

False ## Footnote Relational databases can struggle with aligning schemas when combining data from different sources.

Question 49

Fill in the blank: Continuous maintenance in AI involves ______ to ensure models are effective.

Answer 49

iterating the process ## Footnote Stagnation can lead to outdated models and ineffective performance.

Question 50

What is a potential consequence of improper model maintenance?

Answer 50

Loss of jobs or unfair outcomes, such as incorrect mortgage rates or prison sentences ## Footnote This highlights the importance of responsible AI management.