Machine Learning EMI

上課連結: 週一班 週六班

PART I - Foundations Foundations

Week 1 (slides): 

       Course Logits

  What is Machine Learning?  

  Formal Model

  Lab 0: Data Flow

Week 2:  

       No Lecture (Department administration affair). 

Week 3 (slides):    

      Linear hypothesis class

      Loss and optimization

      Generalization question

      PAC learning

      LAB 1: Learning with a linear classifier

Week 4 (slides):      

     Why finite-class PAC bounds fail for infinite hypothesis classes

     Shattering

     VC dimension

     VC-based generalization and sample complexity

     LAB 2: From theory to practice 

Week 5 (slides):

      No Free Lunch Theorem

      Bias-Variance TradeOff  

Week 6: 

     Convex Learning Problems

     Surrogate Function  

     Stability of Strongly Convex Learning Problems 

Week 7: 

      Spring Holiday (No Lecture)!

Week 8:   

      Midterm I  (9:30~12:00 Close Book. Held in the classroom.)

PART II - Optimization and Sparse Modeling

Week 9: 

      Support Vector Machine

      Kernel Trick

      The Representer Theorem

      KNN

     Decision Tree

     (自主學習) Implementation: Kernel SVM, KNN, Decision Tree, and Random Forest 

 Week 10:  

     Data Preprocessing

     L1 Regularization 

     (自主學習) Implementation: Feature Selection

      Principle Component Analysis

PART III Deep Learning through Optimization

Week 13: 

      LDA

     Kernel PCA

     (自主學習) Implementation: Kernel PCA

     Machine Learning Pipeline and Cross-Validation 

     Learning Curve

     Imbalanced Data 

Week 14: 

     Midterm II 

Week 15: 

     Majority Voting

     (自主學習) Implementation: Majority Voting and Parameter Tuning 

     Bagging      

AdaBoost

     Clustering and K-means

     Determine the number of clusters

     Agglomerative Clustering

     DBSCAN        

Week 16: 

     Final Report

    上課時繳交: Final Exam