|
課程名稱 |
機器學習導論
Introduction to Machine Learning |
|
開課學期 |
113-1 |
|
授課對象 |
重點科技研究學院 奈米工程與科學博士學位學程 |
|
授課教師 |
舒貽忠 |
|
課號 |
AM7196 |
|
課程識別碼 |
543 M1190 |
|
班次 |
|
|
學分 |
3.0 |
|
全/半年 |
半年 |
|
必/選修 |
選修 |
|
上課時間 |
星期三6(13:20~14:10)星期五7,8(14:20~16:20) |
|
上課地點 |
應113應113 |
|
備註 |
總人數上限：100人 |
|
|
|
|
課程簡介影片 |
|
|
核心能力關聯 |
核心能力與課程規劃關聯圖 |
|
課程大綱
|
|
為確保您我的權利,請尊重智慧財產權及不得非法影印
|
|
課程概述 |
本課程著重於機器學習演算法的理解與推導，並強調習題程式撰寫的實作訓練。然而，為了避免因上課時間過於集中而無法充分掌握這些演算法的數學原理，我們設計了以下的課程安排：星期三的課程將以例題為主，幫助學生理解演算法，而星期五的課程則側重於演算法的推導。此外，為了進一步提升學生的程式設計技能，我們還特別安排助教在星期五課程結束後，進行半小時至一小時的習題程式撰寫講解，其中包括Python的教學。具體內容如下：
The course provides a comprehensive introduction to machine learning, with a primary emphasis on the fundamental principles governing learning algorithms. It covers a wide range of topics, including: (1) Supervised Learning: generative and discriminative probabilistic classifiers (Bayes/logistic regression)、least squares regression、Neural Networks (Convolutional Neural Networks, Recurrent Neural Networks)；(2) Probabilistic Graphical Model: Hidden Markov model (HMM)；(3) Basic Learning Theory：PAC learning and model selection. This course aims to provide students with a robust foundation essential for conducting research in machine learning. |
|
課程目標 |
Upon completion, students will be proficient in utilizing calculus, linear algebra, optimization, probability, and statistics to create learning models for diverse real-world challenges. Moreover, they will be well-prepared for advanced research in machine learning and related domains. |
|
課程要求 |
The course is taught in Chinese, utilizing the blackboard for writing and explaining the mathematical principles of machine learning algorithms. 本課程採中文授課，以板書形式，講解機器學習演算法數學原理。 |
|
預期每週課後學習時數 |
|
|
Office Hours |
每週五 17:30~19:30
每週四 15:00~17:00
每週三 15:00~17:00
每週二 15:00~17:00 備註： 陳家宏 R12543002@ntu.edu.tw (週二 15:00-17:00)
王鈺智 R12543051@ntu.edu.tw (週三 15:00-17:00)
鄞振哲 R12543001@ntu.edu.tw (週四 15:00-17:00)
白謹瑜 F10543030@ntu.edu.tw (週五 17:30 - 19:30) |
|
指定閱讀 |
Various readings are required and will be assigned after each lecture. |
|
參考書目 |
1. C. M. Bishop. Pattern Recognition and Machine Learning, Springer, 2006
2. Shai Shalev-Shwartz and Shai Ben-David. Understanding Machine Learning: From Theory to Algorithms, Cambridge University Press, 2014
3. O. Calin. Deep Learning Architectures: A Mathematical Approach, Springer, 2020
4. K. P. Murphy. Probabilistic Machine Learning: An Introduction, MIT Press, 2022
5. Y. S. Abu-Mostafa, M. Magdon-Ismail and H. T. Lin. Learning From Data, AMLbook, 2012
6. E. Alpaydin. Introduction to Machine Learning, MIT Press, 2020 |
|
評量方式
(僅供參考) |
|
|
週次 |
日期 |
單元主題 |
|
第1週 |
9/04,9/06 |
Mathematical formulation of a learning problem, Evaluation of a model (loss function), Generalization error, Empirical Risk Minimization (ERM) |
|
第2週 |
9/11,9/13 |
Bayes optimal classifiers, Overfitting |
|
第3週 |
9/18,9/20 |
Generalization/Empirical errors vs Model complexity, No-Free-Lunch Theorem, Perceptron Learning Algorithm (PLA) |
|
第4週 |
9/25,9/27 |
Perceptron Learning Algorithm (PLA), Probability Review |
|
第5週 |
10/02,10/04 |
Naive Bayes Classifier based on Bernoulli distribution, Maximum Likelihood Estimation (MLE), example of spam classification |
|
第6週 |
10/09,10/11 |
example of classification of digital numbers, Naive Bayes Classifier algorithm, extension to Gaussian distribution, Decision boundary |
|
第7週 |
10/16,10/18 |
Confusion matrix, ROC curve, Logistic Regression |
|
第8週 |
10/23,10/25 |
Sentiment classification example, Logistic Regression, Optimization, Gradient descent |
|
第9週 |
10/30,11/01 |
Gradient descent example, Stochastic gradient descent, Comparison between generative and discriminative models, Neural Networks: abstract neuron, AND, OR and XOR problems |
|
第10週 |
11/06,11/08 |
Multi-layer perception (MLP), mathematical definition, revisit XOR by Boolean operation |
|
第11週 |
11/13,11/15 |
Neural Networks: derive the algorithm of backpropagation |
|
第12週 |
11/20,11/22 |
Explain why BP is efficient, Convolutional Neural Network (CNN), Convolution in 1D and 2D signals, Cross correlation, Equivariance to Translation |
|
第13週 |
11/27,11/29 |
Convolution layer vs fully-connected layer, Characteristics of CNN: sparse and weights sharing, Receptive field, CNN Explainer |
|
第14週 |
12/04,12/06 |
Pooling, CNN architecture, Backpropagation in CNN (derivation) |
|
第15週 |
12/11,12/13 |
Recurrent Neural Networks, or Markov Chains/Hidden Markov Model |
|