2024 spring

기계 학습 / Machine Learning (2분반)

This course covers theoretical backgrounds and practical implementation of different machine learning techniques, including supervised learning, unsupervised learning, deep learning, active learning, anmd reinforcement learning. In addition, it explores the entire pipeline to build applications of machine learning with practices. Furthermore, it provides a broad introduction to ethical issues relevant to machine learning.

Instruction

Course Staff
Time & Location
  • 월/목요일 12:00 - 13:45, 공학 6호관 608호
Office Hours
  • 일시/장소: 화요일 13:00 - 15:00, 공학 6호관 407호
  • 주의 사항
    • 수업 및 과제 관련 내용 질의/응답은 면담 대신 e루리 질의 응답 게시판을 이용
    • 면담 1일전 미리 이메일 연락 등을 통해 일정을 잡을 것
Textbook
  • Primary
    • [Ge23] Aurélien Géron. 2023. Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems, 3rd Ed. O`Reilly
  • Secondary
    • [Oh21] 오일석. 2021. 기계 학습. 한빛 아카데미
    • [Ow22] Louis Owen. 2022. Hyperparameter Tuning with Python: Boost your machine learning model’s performance via hyperparameter tuning. Packt.
    • [Br20] Jason Brownlee. 2020. Data Preparation for Machine Learning, 1.1 Ed. Machine Learning Mastery
    • [Br21] Jason Brownlee. 2021. Imbalanced Classification with Python, 1.3 Ed. Machine Learning Mastery
    • [Mo24] Christoph Molnar. 2024. Interpretable Machine Learning
    • [Ma23] Serg Masis. Interpretable Machine Learning with Python. 2023
Prerequisite
  • (필수) 파이썬프로그래밍
    • 파이썬 사용법 및 Numpy/Pandas 라이브러리 사용법은 숙지된 상태라고 가정하고 수업을 진행
  • (선택) 자료구조, 선형대수학, 데이터분석프로그래밍
Grading Policy
팀 프로젝트Team Project (30%)
  • End-to-End Machine Learning Projects
    • #1 (10%): Data collection
    • #2 (20%): Model building and evaluation
개인 과제Assignments (40%)
  • Kaggle ML Competition
    • #1 (4%): Housing
    • #2 (9%): Wine Quality
    • #3 (9%): Student Dropout or Success
    • #4 (9%): Nurse Stress Prediction using Wearable Sensors
    • #5 (9%): Doom or Animal Crossing
중간 과제Midterm Assignment (20%)
  • Kaggle ML Competition: Classifying Emotions during Debate using Physiological Responses
출석Attendance (10%)
  • 지각 3회 = 결석 1회
  • 결석 1회에 출석 점수 1% 차감
  • 총 수업 일의 1/3 (10회) 초과 결석 시 F
    • 즉, 11회 이상 결석 시 F
  • 별도의 사유(예. 예비군 훈련 등)가 있을 시 수업 시간 전에 교수에게 이메일 송부
    • 단, 급하게 벌어진 사유(예. 급병, 친족상 등)는 소명 자료를 제출

Schedule

W01: Overview
March 04: Overview
March 07: Machine Learning Landscape
  • Lecture
  • Reference
    • [Ge23] Chap. 1
    • [Oh21] Chap. 1
W02: Machine Learning Pipeline
March 11: Machine Learning Pipeline - Lecture
  • Lecture
  • Reference
    • [Ge23] Chap. 2
    • D. Sculley et al. 2015. Hidden technical debt in Machine learning systems. In Proceedings of the 28th International Conference on Neural Information Processing Systems - Volume 2 (NIPS'15).
    • Soowon Kang et al. 2023. K-EmoPhone: A Mobile and Wearable Dataset with In-Situ Emotion, Stress, and Attention Labels. Sci Data 10, 351 (2023).
March 14: Machine Learning Pipeline - End-to-End Practice
W03: Linear Model
March 18: Linear Model - Theory
  • Lecture
  • Reference
    • [Ge23] Chap. 4
    • [Oh21] Chap. 2
March 21: Linear Model - Lab
  • Lab
  • Reference
    • [Ge23] Chap. 4
    • [Oh21] Chap. 2
W04: Support Vector Machine and Decision Tree
March 25: Support Vector Machine
  • Lecture
  • Lab
  • Reference
    • [Ge23] Chap. 5
    • [Oh21] Chap. 11
March 28: Decision Tree
W05: Ensemble Learning
April 01: Basics & Random Forest
  • Lecture
  • Lab
  • Reference
    • [Ge23] Chap. 7
    • [Oh21] Chap. 12
April 04: Gradient Boosting
W06: Feature Engineering #1
April 08: Feature Extraction
  • Lecture
  • Lab
  • Reference
    • Andreas Bulling et al. 2014. A tutorial on human activity recognition using body-worn inertial sensors. ACM Comput. Surv. 46, 3, Article 33.
    • Soujanya Poria et al. 2017. A review of affective computing: From unimodal analysis to multimodal fusion. Information Fusion, 37, 98–125.
April 11: Feature Selection
W07: Feature Engineering #2
April 15: Dimensionality Reduction
  • Lecture
  • Lab
  • Reference
    • [Ge23] Chap. 8
    • [Br20] Chap. 7
April 18: Balancing Label Distribution
  • Lecture
  • Lab
  • Reference
    • [Br21] Chap. 4
W08: Midterm
W09: Cross-Validation and Performance Measures
April 29: Cross-Validation
May 02: Performance Measures
W10: Hyper-parameter Tuning
May 06: Substitution Holiday for Children's Day
  • No class
May 09: Hyper-parameter Tuning #1
  • Lecture
  • Lab
  • Reference
    • [Ow22] Chap. 2, 3, 4, 7, 8
    • Tong Yu and Hong Zhu. 2000. Hyper-Parameter Optimization: A Review of Algorithms and Applications
W11: Clustering
May 13: Hyper-parameter Tuning #2
  • Lecture
  • Lab
  • Reference
    • [Ow22] Chap. 5, 6, 9, 10
    • Tong Yu and Hong Zhu. 2000. Hyper-Parameter Optimization: A Review of Algorithms and Applications
May 16: Clustering
  • Lecture
  • Lab
  • References
    • [Ge23] Chap. 9
    • [Oh21] Chap. 6
W12: Artificial & Deep Neural Network
May 20: Artificial Neural Network
  • Lecture
  • Lab
  • References
    • [Ge23] Chap. 10
    • [Oh21] Chap. 3
May 23: Deep Neural Network
W13: Convolution & Recurrent Neural Network
May 27: Convolution Neural Network
  • Lecture
  • Lab
  • References
    • [Ge23] Chap. 14
    • [Oh21] Chap. 4
May 30: Recurrent Neural Network
  • Lecture
  • Lab
  • References
    • [Ge23] Chap. 15
    • [Oh21] Chap. 8
W14: Autoencoder
June 03: Autoencoder
June 06: Memorial Day
  • No class
W15: Generative Models & Interpretable Machine Learning
June 10: Generative Models
June 13: Interpretable Machine Learning
  • Lecture
  • References
    • [Mo24] Chap. 5 - 10
    • [Ma23] Chap. 4 - 8
W16: Final Term
June 17: Final Term Period
  • No Class; Focus on Team Assignment #2
June 20: Final Remark