| 課程大綱 Syllabus |
學生學習目標
Learning Objectives |
單元學習活動
Learning Activities |
學習成效評量
Evaluation |
備註
Notes |
序
No. | 單元主題
Unit topic |
內容綱要
Content summary |
| 1 | Welcome and introduction to AI |
ntelligent Agents
Application of AI
Terminology
Examples of AI Applications: Computer Games, Self-Driving Car, Machine Translation |
Learning goals: intelligent agents, terminology, and application of AI |
|
|
|
| 2 | Problem Solving (1) |
Definition of a Problem
Route Finding Examples
Tree Search
Graph Search
Breadth-First Search
Depth-First Seach |
Learning goals: Whats is a problem, tree search, graph search, breadth-first search, depth-first search |
|
|
|
| 3 | Problem Solving (2) |
Uniform-Cost Search
Seach Comparison
A* Search
Optimistic Search
State Space
Sliding Block Puzzle |
Learning goals: uniform-cost search, A* search, optimistic search, and state-space |
|
|
|
| 4 | Probability in AI (1) |
Introduction to Probability and Bayes Net
Coin Flip
Dependence/ examples
Bayes Rules
Conditional Independence
Independence
Explain Away
Conditional Dependence
General Bayes Net
D-Separation |
Learning goals: Bayes Rules, Conditional Independence, Independence, Explain Away, Conditional Dependence, General Bayes Net, D-Separation |
|
|
|
| 5 | Probability in AI (2) |
Introduction to Probability and Bayes Net
Coin Flip
Dependence/ examples
Bayes Rules
Conditional Independence
Independence
Explain Away
Conditional Dependence
General Bayes Net
D-Separation |
Learning goals: Bayes Rules, Conditional Independence, Independence, Explain Away, Conditional Dependence, General Bayes Net, D-Separation |
|
|
|
| 6 | Probabilistic Inference |
Enumeration
Variable Elimination
Approximate Inference
Likelihood weighting |
Learning goals: Enumeration, Variable Elimination, Approximate Inference, Likelihood weighting |
|
|
|
| 7 | Machine Learning |
What is ML
Maximum Likelihood
Relationship to Bayes Network
Laplace Smoothing
Linear Regression
Logistic Regression
K-Nearest Neighbors |
Learning goals: Maximum Likelihood, Relationship to Bayes Network, Laplace Smoothing, Linear Regression, Logistic Regression, K-Nearest Neighbors |
|
|
|
| 8 | Representation with Logic |
Propositional Logic
Truth Table
First-Order Logic |
Learning goals: Propositional Logic, Truth Table, First-Order Logic |
|
|
|
| 9 | 期中考 |
期中考 |
期中考 |
|
|
|
| 10 | Planning |
Problem Solving vs. Planning
Planning vs. Execution
Finding a Successful Plan
Classic Planning
Progression Search
Regression Search
Plan Space Search
Situation Calculus |
Learning goals: Problem Solving vs. Planning, Planning vs. Execution, Finding a Successful Plan, Classic Planning, Progression Search, Regression Search, Plan Space Search, Situation Calculus |
|
|
|
| 11 | Introduction to Deep Learning (1) |
What is a Neural Network
Supervised Learning with NN
Why is DL taking off?
Binary Classification
Logistic Regression
Logistic Regression Cost Function
Gradient Descent
Derivative |
Learning goals: Supervised Learning with NN, Binary Classification,
Logistic Regression, Logistic Regression Cost Function, Gradient Descent, Derivative |
|
|
|
| 12 | Introduction to Deep Learning (2) |
What is a Neural Network
Supervised Learning with NN
Why is DL taking off?
Binary Classification
Logistic Regression
Logistic Regression Cost Function
Gradient Descent
Derivative |
Learning goals: Supervised Learning with NN, Binary Classification,
Logistic Regression, Logistic Regression Cost Function, Gradient Descent, Derivative |
|
|
|
| 13 | Deep Learning in Computer Vision (1) |
What Computer “See”
Learning Visual Features
Feature Extraction and Convolution
Convolutional Neural Network (CNN)
An Architecture for Many Applications
MNIST Datasets
Python Programming for CNN |
Learning goals: Visual Features Extraction and Convolution,
Convolutional Neural Network, An Architecture for Many Applications, MNIST Datasets, Python Programming for CNN |
|
|
|
| 14 | Deep Learning in Computer Vision (2) |
What Computer “See”
Learning Visual Features
Feature Extraction and Convolution
Convolutional Neural Network (CNN)
An Architecture for Many Applications
MNIST Datasets
Python Programming for CNN |
Learning goals: Visual Features Extraction and Convolution,
Convolutional Neural Network, An Architecture for Many Applications, MNIST Datasets, Python Programming for CNN |
|
|
|
| 15 | Transfer Learning (1) |
Transfer Learning Concept
MobileNet
TensorFlow Hub and Transfer Learning
Use a TensorFlow Hub MobileNet for prediction
Use a TensorFlow Hub model for the Cats vs. Dogs dataset
Do simple transfer learning with TensorFlow Hub |
Learning goal: Transfer Learning Concept, MobileNet, Use a TensorFlow Hub MobileNet for prediction, Use a TensorFlow Hub models for the Cats vs. Dogs dataset, Do simple transfer learning with TensorFlow Hub |
|
|
|
| 16 | Transfer Learning (2) |
Transfer Learning Concept
MobileNet
TensorFlow Hub and Transfer Learning
Use a TensorFlow Hub MobileNet for prediction
Use a TensorFlow Hub model for the Cats vs. Dogs dataset
Do simple transfer learning with TensorFlow Hub |
Learning goal: Transfer Learning Concept, MobileNet, Use a TensorFlow Hub MobileNet for prediction, Use a TensorFlow Hub models for the Cats vs. Dogs dataset, Do simple transfer learning with TensorFlow Hub |
|
|
|
| 17 | Time-Series Forecasting |
Applications
Common Patterns
Forecasting
Metrics
Time Windows
Forecasting with Machine Learning
Forecasting with an RNN |
Learning goal: Applications, Common Patterns, Forecasting, Metrics, Time Windows, Forecasting with Machine Learning, Forecasting with an RNN |
|
|
|
| 18 | 期末考 |
期末考 |
期末考 |
|
|
|