Time: Fri 1:00-4:00pm

Location: Cluster Bd. R509 (학연산클러스터 509호)

Textbook: Introduction to Statistical Learning with Applications in R, James, Witten, Hastie & Tibshirani, Springer (free PDF)

Grading:

• • Homework: 60%

  • Project: 20%

  • Attendance: 10%

  • Discussion: 10%

Project

• • 5.25: 정상수업 (강의실 변경: 학연산 3층 304)

  • 6.1: 휴강 (비디오 강의로 대체)

  • 6.8: 프로젝트 발표 (5분) : 최종발표 형식으로 준비하되, 결과 부분만 발표시점까지 결과를 발표하면 됩니다.

    • Motivation

    • Problem description

    • Result

  • 6.15: 휴강 (비디오 강의로 대체)

  • 6.22: Final Report (due by 13pm, 학연산 620호, 12 페이지)

    • 프로젝트 발표 내용 +

      • Method 설명

      • Result + discussion

      • Reference

      • Reproducibility (optional: github repository of experimentation scripts)

    • 과제 제출 (컴퓨터로 작성하여 출력 요망)

      • HW4: 4.7 Exercises #4, #6

      • HW5: 9.7 Exercise #5

• Submit Final Report

TA e-mail : "dongykang@gmail.com"

• Lecture Videos by the Authors: Link

• R Tutorial (pdf)

• Understanding R regression plot [link]

Project data suggetions:

• UCI Machine Learning Repository

• Kaggle

• 공공데이터포털

• data.gov

Lecture Notes

• 01. Introduction (PDF)

• 02. Statistical Learning (PDF)

• 03. Linear Regression (PDF):

  • Univariate linear regression (slides ~ p13) + R introduction (code)

  • HW1 (due Mar 23): 2.4 Exercises #1, #2, #4, #7, #8, #9

• 04. Linear Regression

  • Multivariate linear regression (slides ~ end)

• 05. Classification (PDF, slides ~ p4)

  • Lab: linear regression (code)

  • HW2 (due Apr 6): 3.7 Exercises #5, #9

• 06. Classification (slides ~ p18)

  • Logistic regression

  • HW3 (due Apr 11): 3.7 Exercises #10, #13

• 07. Classification (slides ~ p32)

  • LDA , QDA

• 08. Resampling methods (PDF)

• 09. Linear Model Selection and Regularization (PDF)

  • Subset selection, Ridge regression, LASSO

• 10. Linear Model Selection and Regularization

  • PCR, PLS

  • R Lab (code)

  • HW4 (due May 18): 4.7 Exerciese #4, #6

• 11. Moving beyond linearity (PDF)

  • R Lab (code)

  • HW discussion: 3.7 Exercises #10, #13 (이기찬)

• 12. Tree-based Method (video)

• 13. SVM (video1, video2)

• HW Submissions (due by June 22, 1pm, 학연산 620호)

  • • HW4: 4.7 Exercises #4, #6

    • HW5: 9.7 Exercise #5