機器學習

臺北科技大學 115-1 學期課程，課號 364898。

課程大綱

本課程以深入淺出之方式，介紹與機器學習之相關技術及演算法，目標是修習本課程之學生具有閱讀機器學習相關文獻之基本知識，並能從事相關之實務或電腦模擬實驗之工作，因此本課程著重在廣泛介紹各類之演算法之應用，而非深究少數演算法之深奧數學原理．本課程涵蓋之演算法，包含：最佳化基本知識，隨機最佳化演算法，隨機搜尋，決策樹，隨機森林，類神經網路及深度學習相關演算法，基因演算法，貝氏定理及應用，高斯混合模型，隱藏式馬可夫鍊(及訓練演算法)，支援向量機，線性及邏輯回歸，分群方法，降維方法，提升方法，決策融合技巧，及增強式學習等。 This course covers most of the techniques and algorithms related to the current trend in the machine learning field. The goal of the course is that students complete this course can read and understand the literature in the machine learning field, and can conduct computer simulations for experiments or apply the techniques to practical problems. Therefore, the course intends to cover a broad range of algorithms with emphasis on applications, rather than to cover only a few algorithms with extensive mathematical treatments. The algorithms covered in this course include: foundations of optimizations, random search, decision tree, random decision forests, neural networks and deep learning, genetic algorithm, Bayes' theorem and its applications, Gauss mixture model, hidden Markov model, support vector machine, linear and logistic regressions, clustering algorithms, dimension reduction techniques, boosting methods, decision level fusion techniques, and reinforcement learning.

課程進度

Week 1: Class Announcement Week 2: Introduction to Machine Learning Week 3: Basics of Supervised Learning (Classification and Regression) Week 4: VC dimension, Bayesian Decision Theory, Naive Bayes Classifiers Week 4: Dimension Reduction Techniques: PCA, FA, LDA, ICA Week 5: Unsupervised learning: Clustering Algorithm (k-means) Week 6: Decision Trees, Bagging Algorithm (Random Forest) Week 7: Basics of optimization Week 8: Basics of Optimization Week 9: Gradient Search Week 10: Feedforward Neural Networks Week 11: Multi-layer Perceptrons, Back Propagation, Regularization Methods Week 12: Deep Learning and Convolutional Neural Networks Week 13: Deep Neural Networks: YOLO, Autoencoder, LSTM Week 14: Bayesian Networks, Confidence of Classification, Self-supervised learning Week 15: SVM, Adaboost, Ensemble Learning Week 16: Final exam Week 17: Presentation Week 18: Presentation

評量方式

• Programming Assignments • Exams • Class Participation

教材

【遵守智慧財產權觀念，請使用正版教科書，不得使用非法影印教科書】 使用外文原文書：否 Course PPT files are available for students.