CO6062: Machine Learning Algorithms, Fall 2024
Lecturer:
Prof. Dah-Chung Chang, E-mail: dcchang@ce.ncu.edu.tw
Office: E1-311, TEL: 03-4227151 ext. 35511
Lecture Time:
Tuesday 1:00pm-4:00pm
TA Information:
E1-306, TEL:
03-4227151 ext 35534
Pre-requisite:
Programming,
Linear Algebra, Calculus, Probability Theory, (Adaptive Filtering Theory)
Course Goal:
You can learn main knowledge of machine learning
and deep learning algorithms and their programming skills with MATLAB and
Python/Pytorch on Google’s TensorFlow such that you
can apply them to the fields you are studying or will study. Please group your
team for the final project at most
two persons.
Course Outline:
Part A: Supervised Learning
Least Squares and
Linear Regression Classifications
Linear Discriminant Analysis (LDA)
Bayesian Classification
Logistic Regression (LR)
Decision Tree
Support Vector Machine (SVM)
Kernel Method
MATLAB Exercises
Part B: Unsupervised Learning
Clustering
K-Means
Principal Component
Analysis (PCA)
butterfly.gif
Singular Value
Decomposition (SVD)
Independent
Component Analysis (ICA)
MATLAB Exercises
Part C: Deep Learning
Multilayer
Perceptron (MLP)
Deep Neural
Networks (DNN)
Convolutional
Neural Networks (CNN)
Recurrent Neural
Networks (RNN)
Long Short-Term
Memory (LSTM)
Python/Pytorch Tutorial Lab./Exercises
Part D: Advanced AI Topics
Generative
Adversarial Network
Reinforcement
Learning
Autoencoder
Ensemble
Learning
Homeworks Assignment:
* Note:
Each team needs to email your homeworks along
with codes for score evaluation. If no figures and codes are attached to your
reports, we cannot judge the validity.
Notice: Upload will be automatically closed after
2 days over the deadline.
|
Items |
Deadline |
Subject |
Upload
web site |
|
HW#1 |
Comparison of Classification and Clustering Methods |
||
|
HW#2 |
ICA for Blind Source Separation |
||
|
HW#3 |
TA_Lab1: SVM with Python (See Lecture notes) TA ML Python Coding Tutorial |
||
|
HW#4 |
TA_Lab2: CNN and RNN with Python (See Lecture notes) TA DL Python Coding Tutorial |
||
|
HW#5 |
Autoencoders |
||
|
Team Report |
PPT report and
discussion for Team Project plan (10 mins for each team) |
||
|
Final Project |
Final Project Report
Delivered |
Final Project:
Final
Project Deadline:
Final
Project Presentation (max. 10 mins/team):
-
You SHALL
list the homework/project title and member names/IDs (one person for all homeworks) on the first page of your reports, and upload your homeworks, reports, and final project documents along
with MATLAB/Python/Pytorch codes in the .doc format (Word) and the
Presentation Report (PPT) to the upload web sites.
Notice
that in your word .doc report, the detailed materials about your problem,
system model, algorithms, simulation results, references, and codes SHALL be
described and explained!
The uploading date should be
not late than the deadline.
Course Materials
Textbook:
Lecture
Notes
References:
See them in Lecture Notes
Grading
1.
Policy: Homeworks*75%+Final Project*20%+Team Project Discussion*5%
2.
Term Grading Report
