Chapter 01 Introduction to Artificial Intelligence

01 The Age of Artificial Intelligence
The impact of artificial intelligence
Artificial intelligence permeates our lives
Artificial intelligence and humans
02 Definition of Artificial Intelligence
intelligent agent
Artificial Intelligence vs. Machine Learning vs. Deep Learning
03 Turing Test
ELIZA
The Chinese Room
Eugene Goostman in 2014
Problems with the Turing Test
04 History of Artificial Intelligence
The Birth of Artificial Intelligence (1943-1956)
Golden Age (1956-1974)
The First AI Winter (1974-1980)
Golden Age (1980-1987)
The Second AI Winter (1987–1993)
The Resurgence of Artificial Intelligence (1993-2011)
Deep Learning, Big Data, and Artificial Intelligence (2011-present)
05 Where is artificial intelligence needed?
self-driving cars
Video recommendation system
advertising system
Chatbot
Medical field
Art creation
New drug development and biology
Super-large AI
Mini Project: Trying Out Google's Deep Dream
Try Mini Project ChatGPT
Summary
Practice problems

Chapter 02 Exploration

01 Explore
02 State space exploration problem
State space exploration problem
LAB path finding problem
LAB N-queen problem
03 Navigation Tree
LAB 4-queen problem search tree
04 Basic Search Techniques
Measuring navigation performance
05 Depth-first search
Analysis of depth-first search
06 Breadth-first search
Analysis of breadth-first search
07 Depth-limited exploration
Pros and Cons of IDDFS
08 FS and DFS 8-puzzle programs
How to represent the board?
What will be used to implement open and closed queues?
How to create child nodes?
BFS full source code
DFS program
09 Empirical Exploration Methods
10 Hill Climbing Techniques
Algorithm
Local biggest problem
11 Top Priority Search
12 A* algorithm
LAB A* algorithm simulation
13 Python implementation of the A* algorithm
Search algorithm for the LAB N-queen problem
Mini Project TSP
Summary
Practice problems

Chapter 03 Game Tree

01 Game Program
Game definition
Game tree for Tic-Tac-Toe
02 Minimax Algorithm
Applying the Minimax Algorithm to the Tic-Tac-Toe Game
LAB Minimax Algorithm Practice
Pseudocode of the minimax algorithm
Minimax Performance Analysis
03 Tic-Tac-Toe Game Programming
04 Alphabeta Pruning
AlphaBeta algorithm
AlphaBeta Algorithm Practice
05 Incomplete Decision
Summary
Practice problems

Chapter 04 Expert Systems

01 Expert System
History of expert systems
02 Components of an expert system
Knowledge Base
Inference Engine
User interface
03 Knowledge and Artificial Intelligence
Data, information, knowledge
Rule 04
You can use AND or OR in rules.
05 Inference in Expert Systems
Forward inference
backward reasoning
LAB Inference Practice
LAB Fire Treatment System
06 Conflict Resolution
07 Advantages and Disadvantages of Expert Systems
Summary
Practice problems

Chapter 05 Knowledge Representation

01 Knowledge Representation
Rule 02
03 Semantic Network
04 Frame
Advantages of Frames
Frames and Object-Oriented Programming
Frames and Inheritance
Semantic Web and Frames
05 Logic
06 Propositional logic
Inference in propositional logic
Modus Ponens
Modus Tollens
Syllogism
07 Predicate logic
08 Inference in predicate logic
Jeong Hyeong-sik
Logic Fusion (Resolution)
Proof by logical fusion
09 Introduction to the Semantic Web and Ontology
Problems with the existing web
Semantic Web
Examples of the Semantic Web
Elements of Semantic Web Technology
Ontology
10 Prologue
LAB Logic Convergence Lab #1
LAB Logic Convergence Practice #2
LAB Logic Convergence Practice #3
Summary
Practice problems

Chapter 06 Fuzzy Logic

01 What is fuzzy logic?
Areas where fuzzy logic can be used
02 Crisp sets and fuzzy sets
Crisp set
fuzzy sets
Notation of fuzzy sets
Example of LAB fuzzy sets
03 Operators in fuzzy sets
Centralized operators CON and DIL
LAB fuzzy set operator
04 Fuzzy Inference
fuzzy rules
Fundamentals of Fuzzy Inference
The process of fuzzy inference
If there are multiple rules
Problems giving LAB tips
Summary
Practice problems

Chapter 07 Uncertainty

01 Uncertainty
Examples of uncertainty
Why does uncertainty arise?
Handling Uncertainty in Artificial Intelligence Systems
02 Handling uncertainty using probability
Prior and posterior probabilities
Bayes' theorem
LAB Could I be infected with the Z-virus?
LAB Fruit Problem
LAB Card Game
03 Bayes' Theorem and Inference
When there is multiple evidence and hypotheses
Disadvantages of Bayes' theorem
Calculating the probability of a rule using LAB Bayes' theorem
LAB Spam Filtering
LAB Monty Hall Problem
04 Confidence
Definition of certainty
Confidence in rules with uncertain evidence
When a rule has multiple premises
LAB Confidence Practice
Summary
Practice problems

Chapter 08 Genetic Algorithms

01 Evolution in Nature
02 Genetic Algorithm
Chromosomes, encodings, and evaluation functions
Pseudocode for genetic algorithm
Selection operator
Crossover operator
mutation operator
genetic algorithm
03 Genetic Algorithm Example
04 Genetic Algorithm Program
LAB 8-queen problem
LAB TSP Problem
05 Advantages and Disadvantages of Genetic Algorithms
06 Genetic Programming
How to express the program?
Basic operations
GP algorithm
Summary
Practice problems

Chapter 09: Introduction to Machine Learning

01 What is machine learning?
Differences Between Machine Learning and Traditional Programming
Artificial intelligence, machine learning, deep learning
History of Machine Learning
Types of machine learning
02 Machine Learning Terminology
Feature
Function Approximation and Machine Learning
Label
sample
Learning and Prediction
Training data and test data
03 Supervised Learning
Regression
Classification
Mini Project: Experience machine learning using teachable machines.
04 Unsupervised Learning
05 Reinforcement Learning
06 Machine Learning Process
Data collection
Training data and test data
Model selection
learning
evaluation
prediction
07 Performance Evaluation of Machine Learning Algorithms
Accuracy
confusion matrix
Mini Project: Experience Machine Learning
08 Uses of Machine Learning
Where is machine learning used?
The Practical Value of Machine Learning for Programmers
Summary
Practice problems

Chapter 10 Linear Regression

01 Linear Regression
Introduction to Linear Regression
Types of linear regression
Principles of linear regression
Learning and Loss
02 How to minimize the loss function in linear regression
Analytical method
Gradient Descent Method
Gradient descent in linear regression
03 Linear Regression Python Implementation #1
04 Linear Regression Python Implementation #2
Let's graph linear regression
LAB Linear Regression Practice
05 Overfitting vs. Underfitting
LAB Diabetes Example
Predicting house prices based on LAB area
Summary
Practice problems

Chapter 11 kNN, K-means, and Decision Trees

01 kNN algorithm
Summary of the kNN algorithm
Modified kNN algorithm
Advantages and Disadvantages of the kNN Algorithm
02 Example: Iris Classification Using kNN
Features and Labels
Drawing a graph
Learning kNN
Let's predict
03 Example: Classifying Handwritten Images with kNN
Reading the dataset
Training data and test data
model
learning
Prediction and Evaluation
04 Performance Evaluation of Machine Learning Algorithms
Let's print the confusion matrix
Classification Report
05 K-means clustering
Example of K-means clustering
K-means clustering
06 K-means clustering using sklearn
Include the library
Prepare the data
Data visualization
Create a cluster
How to determine k
Implementation of the elbow method
LAB K-means Clustering Practice
07 Decision Tree
Principles of decision trees
How do you build a tree?
entropy
Decision trees using sklearn
Example 08: Classifying Irises Using Decision Trees
Summary
Practice problems

Chapter 12 Perceptron

01 Neural Network
Advantages of neural networks
Mathematical model of neurons
02 Perceptron
Can perceptrons learn logical operations?
03 Perceptron Learning Algorithm
Example
Practicing Perceptrons with Sklearn
04 Limitations of the Perceptron
Learning XOR operation
Linearly classifiable problems
Solving the XOR problem with a multilayer perceptron
Classified as LAB perceptron
Summary
Practice problems

Chapter 13 Multilayer Perceptron (MLP)

01 Multilayer Perceptron
Activation function
step function
sigmoid function
ReLU function
Hyperbolic tangent function
02 Forward Pass
Let's express it as a matrix
03 Error calculation
What is a loss function?
A concrete example of a loss function
Example of calculating a loss function
04 Reverse Pass
What do you need to know?
gradient descent
Reintroducing Gradient Descent
gradient
LAB Gradient Descent Practice
05 Backpropagation Algorithm
Application of chain rules
Hidden layer nodes
Let's conclude
Backpropagation learning algorithm
LAB Backpropagation Algorithm Animation
06 MLP Implementation Using NumPy
1) Forward propagation
2) Perform error backpropagation
3) Write a function to test after learning is complete.
4) Call the training function and test function sequentially.
07 Practice using Google Playground
Epoch
learning rate
Choosing an activation function
Problem type
Ratio of training data to test data
Select input features
Adding a hidden layer
Start learning
Classification practice without hidden layers
Practice with added hidden layer
08 Google's TensorFlow
Installing TensorFlow
Keras examples
MNIST Digit Recognition Using LAB MLP
Summary
Practice problems

Chapter 14: Deep Neural Networks and Deep Learning

01 Deep Learning
The role of the hidden layer
02 Vanishing gradient problem
New activation function
Mini Project Activation Function Experiment
03 Loss function problem
Mean square error
Softmax activation function
Cross entropy loss function
Calculating LAB cross entropy
04 Loss Functions in Keras
BinaryCrossentropy
CategoricalCrossentropy
SparseCategoricalCrossentropy
MeanSquaredError
05 Weight Initialization Problem
Weight initialization method
Mini Project Weight Initialization Experiment
06 Mini Batch
07 Data Normalization
08 Data Encoding Techniques
09 Learning rate and momentum
10 Handling overfitting
How to detect overfitting
How to handle overfitting
Early termination
Weight regulation method
Data Augmentation
Dropout
ensemble
Mini Project Batch Size, Learning Rate, Regularization Term
11 Example: MNIST Digit Recognition
Import numeric data
Building a Model
Teaching
Example 12: MNIST Fashion Item Classification
Using a fully connected neural network
Summary
Practice problems

Chapter 15 Convolutional Neural Networks

01 Introduction to Convolutional Neural Networks
The Origins of Convolutional Neural Networks
Neocognitron
The Importance of Convolutional Neural Networks
Mini Project: Experience Convolutional Networks
02 Convolution operation
Structure of a convolutional neural network
Convolution operation in image processing
Convolution operations in convolutional neural networks
stride
padding
Number of kernels
03 Pooling (subsampling)
04 Example: MNIST Fashion Item Classification
Summary
Practice problems

Chapter 16 Image Recognition

01 What is image recognition?
Mini Project: Experience Image Recognition Neural Networks
02 Traditional image recognition
03 Image recognition using deep neural networks
04 Example: Classifying CIFAR-10 Images
CIFAR-10 dataset
CIFAR-10 classification program using convolutional neural networks
05 Data Augmentation
Example 06: Distinguishing between dogs and cats
Dog and Cat Dataset
Installing the library
Image output
Creating a neural network model
Image preprocessing
learning
Display learning results graph
07 Weight Storage and Transfer Learning
Saving and loading learned weights
transfer learning
Pre-trained neural network model
Redefine a pretrained model for your project
Example #1: Writing a Dog Recognition Program
Example #2: Using a pretrained model as a feature extractor preprocessor.
Summary
Practice problems

Chapter 17 Reinforcement Learning

01 What is reinforcement learning?
Principles of reinforcement learning
Reinforcement Learning Framework
Reinforcement Learning and Deep Learning
02 Ice Lake Game
OpenAI Foundation
Frozen Lake game
Ice Lake Game Version #1
Analysis of execution results
03 Q-Learning #1
compensation
Q-function
Policy
Q-value circular relationship
Let's actually calculate the Q-value in the ice lake problem.
Ice Lake Game Version #2
04 Q-Learning #2
Exploration and Exploitation
Discounted rewards
learning rate
Ice Lake Game Version #3
05 Deep Q-Learning
Why use neural networks?
DQN (Deep Q Network)
Q-Learning vs. Deep Q-Learning
How to learn
Algorithm
Practical application examples
Problem
Disadvantages of Deep Q-Learning
Summary
Practice problems

Chapter 18 Generative Models

01 What is a generative model?
Generative models identify rules for generating training data.
Differences between discriminative and generative models
02 What is GAN?
GAN structure
GAN training process
Discriminator training
Generator training
loss function
03 Example: Generating Digit Images with GAN
Summary
Practice problems

Search

The 2016 Go event in which AlphaGo defeated Lee Sedol shocked people around the world, and artificial intelligence became one of the most exciting fields in computer science.
Since the development of deep learning algorithms and the objective demonstration of their superiority in an image recognition competition held in 2012, almost every country has been developing artificial intelligence technology as a core technology of the future.
Having overcome the AI ​​winter that has occurred twice in the history of AI, a new era of AI is dawning, based on the science of extensive big data collection and analysis, highly advanced semiconductor integration, and ultra-high-speed processing technology.

This book is a must-read for those preparing for the new era of artificial intelligence. It allows both engineering students majoring in convergence technologies such as computers, electronics, information and communication, and machinery, as well as non-computer science majors new to artificial intelligence, to learn the basic concepts and theories of artificial intelligence through various examples and practical exercises.
It covers all the fundamental theories of artificial intelligence and provides detailed explanations of machine learning, including deep learning, across 10 chapters.
In the field of deep learning, you can practice using Keras, a high-level library based on Google's TensorFlow, and experience it using Google's TensorFlow Playground.