中正大學課程大綱
課程名稱(中文)： 機器學習 開課單位： 資訊工程研究所(Graduate Institute of Computer Science and Information Engineering)
課程名稱(英文) Machine Learning 課程代碼 4105931_01
授課教師： 江振國 學分數 3
必/選修 選修 開課年級 碩博合開
先修科目或先備能力：
This is an introductory course, intended for senior undergraduate and graduate students. The prerequisites are courses on computer programming, algorithms, probability, and linear algebra.
課程概述：
Machine learning is programming computers to optimize a performance criteria using example data or past experience. In this course, we will study key algorithms and theory that form the core of machine learning such as conpcet learning, decision trees, supervised learning, unsupervised learning, bayesian lerning, neural networks, and support vector machines. Students are expected to learn their problem setting, algorithms, and assumptions that underline each.
學習目標：
1. learn how to formulate, experiment and measure a machine learning algorithm.
2. learn core allgorithms and theory of machine learning.
3. learn how to solve applications by machine learning.
教科書：
Machine Learning, Tom. Mtichell, McGraw-Hill

課程大綱 分配時數 核心能力 備註
單元主題 內容綱要 講授 示範 隨堂作業 其他
Introduction
Introduction to Machine Learning
3 12345678
Learning Problems and Types
Applications of Machine Learning,
Components of Machine Learning,
Machine Learning Types
4 1 12345678
Overfitting and Validation
Overfitting Problem, The Role of Noises and Data size, Leave-One-Out and K-fold Cross Validation
 4 1 12345678
Linear Regression and Regularization
Linear Regression Problem and Algorithm. Generalization, Regularization
3 1 12345678
Tree-based Learning and Boosting Methods
Decision Tree, Random Forest, XGBoost, Extra Regression Tree and Adaptive Boosting
 4 1 12345678
Clustering and Association Discovery
k-Mean Clustering, EM Algorithm, Hierarchical Custering, Association Rules
4 1 12345678
Support Vector Machines
Hyperplane Classifiers, Optimal Separating Hyperplane, Kernel Functions, Support Vector Machine
4 2 12345678
Neural Networks
Perceptron Learning, Multilayer Network and Backpropagation Algorithm
 5 1 12345678
Deep Learning
Convolutional Neural Network, Recurrent Neural Network, LSTM and Generative Adversarial Network
 6 1 12345678
Applications of Machine Learning
Text Categorization, Information Extraction, and Computer Vision
4 1 12345678

教育目標
1.具獨立從事學術研究或產品創新研發之人才
2.具團隊合作精神及科技整合能力，並在團隊中扮演領導、規劃、管理之角色
3.具自我挑戰與終身學習能力之人才
4.具有學術倫理、工程倫理、國際觀之人才

核心能力
1.具有資訊工程與科學領域之專業知識(Competence in computer science and computer engineering.)
2.具有創新思考、問題解決、獨立研究之能力(Be creative and be able to solve problems and to perform independent research.)
3.具有撰寫中英文專業論文及簡報之能力(Demonstrate good written, oral, and communication skills, in both Chinese and English.)
4.具策劃及執行專題研究之能力(Be able to plan and execute projects.)
5.具有溝通、協調、整合及進行跨領域團隊合作之能力(Have communication, coordination, integration skills and teamwork in multi-disciplinary settings.)
6.具有終身學習與因應資訊科技快速變遷之能力(Recognize the need for, and have the ability to engage in independent and life-long learning.)
7.認識並遵循學術與工程倫理(Understand and commit to academic and professional ethics.)
8.具國際觀及科技前瞻視野(Have international view and vision of future technology.)

請尊重智慧財產權，不得非法影印教師指定之教科書籍

教學要點概述：
1. 教材編選(可複選)：自編簡報(ppt)教科書作者提供
2. 教學方法(可複選)：講述板書講述
3. 評量工具(可複選)：上課點名 10.00%, 隨堂測驗0%, 隨堂作業30.00%, 程式實作0%, 實習報告0%,
                       專案報告0%, 期中考30.00%, 期末考0%, 期末報告30.00%, 其他0%,
4. 教學資源：課程網站 教材電子檔供下載 實習網站
5. 教學相關配合事項：

課程目標與教育核心能力相關性        
請勾選：12345678
1 具有資訊工程與科學領域之專業知識(Competence in computer science and computer engineering.)
為何有關：
Machine Learning has been sucessfully applied to many of the recent applications such as data mining, information filtering and extraction, robotics and bioinformatics. It has become one of the main approaches for solving problems by computers.
達成指標：
knowledge of fundamental machine techniques and their applications.
評量工具(可複選)：
Assignments, examinations, and the final report.
2 具有創新思考、問題解決、獨立研究之能力(Be creative and be able to solve problems and to perform independent research.)
為何有關：
student will learn how to formulate a complex problem as a machine learning problem, and develop an efficient program to solve the problem
達成指標：
the capability of solving complex problems by machine learning techniques
評量工具(可複選)：
Assignments, Exams, and Final Report
Level 5: Submitted 80% of assignments or expected final grade is 80 or above or report score is 80 or above.
Level 4: Submitted 60% of assignments or expected final grade is 70 or above or report score is 70 or above.
Level 3: Submitted 40% of assignments or expected final grade is 60 or above or report score is 60 or above.
Level 2: Submitted 20% of assignments or expected final grade is 50 or above or report score is 50 or above.
Level 1: Did not submit assignments or expected final grade is below 50 or report score is below 50.
6 具有終身學習與因應資訊科技快速變遷之能力(Recognize the need for, and have the ability to engage in independent and life-long learning.)
為何有關：
Machine Learning has been sucessfully applied to many of the recent applications, and has become one of the main approaches for solving problems by computers.
達成指標：
the capability of reading research papers on recent development of machine learning
評量工具(可複選)：
Final Report