Machine Learning

1 Learning Path

1.1 Phase 1: Foundations (2-3개월, 수학적 기초와 ML 핵심 개념 확립)

Mathematical Prerequisites - Linear Algebra ⭐ - Calculus and Optimization ⭐ - Probability and Statistics ⭐

ML Fundamentals - ML 개념과 학습 유형 - Bias-Variance Tradeoff ⭐ - Data Preprocessing - Feature Engineering ⭐

Basic Algorithms - Linear Regression ⭐ - Logistic Regression ⭐ - Decision Trees ⭐

1.2 Historical Methods

• 2022-12-09, PCA (Principal Component Analysis)
• 2023-02-03, PyTorch Introduction
• 2023-02-03, TensorFlow Introduction
• 2023-02-06, Naive Bayes
• 2023-03-14, 산술·기하·조화평균의 직관적 이해

1.3 Phase 2: Core Algorithms (2-3개월, 주요 알고리즘)

Classification & Regression - SVM ⭐ - KNN - Naive Bayes - Regularization ⭐

Ensemble Methods - Random Forest ⭐ - Gradient Boosting ⭐ - XGBoost ⭐ - LightGBM ⭐

Unsupervised Learning - K-Means ⭐ - Hierarchical Clustering - PCA ⭐ - DBSCAN

1.4 Phase 3: Advanced Methods (2-3개월, 고급 기법)

Model Evaluation - Cross-Validation ⭐ - Performance Metrics ⭐ - Hyperparameter Tuning ⭐

Advanced Topics - Imbalanced Learning ⭐ - Gaussian Processes - Model Interpretability (SHAP) ⭐

Special Applications - Time Series - NLP Basics - Recommender Systems

1.5 Phase 4: ML Engineering (2개월, 프로덕션 배포와 운영)

Production ML - Model Serving - Model Monitoring ⭐ - A/B Testing for Models

Scalability - Distributed Training - Model Compression - Feature Stores

2 Implementation Notes

2.1 Python Libraries

Interpretability - import shap ⭐ - import lime

Production - import mlflow ⭐ - import bentoml

2.2 Recommended Project Structure

ml_project/ ├── data/ │ ├── raw/ │ ├── processed/ │ └── features/ ├── notebooks/ │ ├── eda.ipynb │ ├── modeling.ipynb │ └── evaluation.ipynb ├── src/ │ ├── preprocessing/ │ ├── features/ │ ├── models/ │ └── evaluation/ ├── tests/ ├── configs/ └── models/

3 Foundations

3.1 Mathematical Prerequisites

• Linear Algebra for ML (ML을 위한 선형대수)
  • 1111-11-11, Vectors and Matrices (벡터와 행렬)
  • 1111-11-11, Matrix Operations (행렬 연산)
  • 1111-11-11, Eigenvalues and Eigenvectors (고유값과 고유벡터)
  • 1111-11-11, Singular Value Decomposition (특이값 분해)
• Calculus and Optimization (미적분과 최적화)
  • 2026-03-23, 함수의 합 vs 합성함수 — Sequential vs Joint 추정
  • 1111-11-11, Derivatives and Gradients (도함수와 기울기)
  • 1111-11-11, Chain Rule and Backpropagation (연쇄법칙과 역전파)
  • 1111-11-11, Convex Optimization (볼록 최적화)
  • 1111-11-11, Gradient Descent Methods (경사하강법)
• Probability and Statistics (확률과 통계)
  • 1111-11-11, Probability Distributions (확률분포)
  • 1111-11-11, Maximum Likelihood Estimation (최대우도추정)
  • 1111-11-11, Bayesian Inference (베이지안 추론)
  • 1111-11-11, Statistical Testing (통계적 검정)

3.2 ML Fundamentals

• Introduction to Machine Learning (머신러닝 소개)
  • 1111-11-11, What is Machine Learning? (머신러닝이란?)
  • 1111-11-11, Types of Learning (학습의 종류)
    • Supervised Learning (지도학습)
    • Unsupervised Learning (비지도학습)
    • Semi-supervised Learning (준지도학습)
    • Reinforcement Learning (강화학습)
  • 1111-11-11, Bias-Variance Tradeoff (편향-분산 트레이드오프) ⭐
  • 1111-11-11, Overfitting and Underfitting (과적합과 과소적합)
  • 1111-11-11, No Free Lunch Theorem (공짜 점심 정리)
• Data Preprocessing (데이터 전처리)
  • 1111-11-11, Data Cleaning (데이터 정제)
    • Missing Value Handling (결측치 처리)
    • Outlier Detection (이상치 탐지)
    • Data Quality Assessment (데이터 품질 평가)
  • 1111-11-11, Feature Scaling (특성 스케일링)
    • Standardization (표준화)
    • Normalization (정규화)
    • Robust Scaling (로버스트 스케일링)
  • 1111-11-11, Encoding Categorical Variables (범주형 변수 인코딩)
    • One-Hot Encoding
    • Label Encoding
    • Target Encoding
    • Embeddings
• Feature Engineering (특성 공학) ⭐
  • 1111-11-11, Feature Creation (특성 생성)
    • Polynomial Features (다항 특성)
    • Interaction Features (상호작용 특성)
    • Domain-specific Features (도메인 특화 특성)
  • 1111-11-11, Feature Selection (특성 선택)
    • Filter Methods (필터 방법)
    • Wrapper Methods (래퍼 방법)
    • Embedded Methods (임베디드 방법)
  • 1111-11-11, Dimensionality Reduction (차원 축소)
    • Principal Component Analysis (주성분 분석)
    • Linear Discriminant Analysis (선형판별분석)
    • t-SNE and UMAP

4 Supervised Learning

4.1 Regression (회귀)

• Regression Analysis (회귀 분석)
  • 1111-11-11, Linear Regression (선형 회귀) ⭐
    • Ordinary Least Squares (최소자승법)
    • Geometric Interpretation (기하학적 해석)
    • Assumptions and Diagnostics (가정과 진단)
    • Coefficient Interpretation (계수 해석)
  • 1111-11-11, Regularization (정규화) ⭐
    • Ridge Regression (L2 정규화)
    • Lasso Regression (L1 정규화)
    • Elastic Net (탄력망)
    • Regularization Path (정규화 경로)
  • 1111-11-11, Polynomial Regression (다항 회귀)
  • 1111-11-11, Generalized Linear Models (일반화 선형모형)
    • Logistic Regression (로지스틱 회귀)
    • Poisson Regression (포아송 회귀)
    • Link Functions (연결함수)

4.2 Classification (분류)

• Classification Methods (분류 방법)
  • 1111-11-11, Logistic Regression (로지스틱 회귀) ⭐
    • Binary Classification (이진 분류)
    • Multinomial Logistic Regression (다항 로지스틱)
    • Odds Ratio Interpretation (오즈비 해석)
  • 1111-11-11, Naive Bayes (나이브 베이즈)
    • Gaussian Naive Bayes
    • Multinomial Naive Bayes
    • Bernoulli Naive Bayes
  • 1111-11-11, K-Nearest Neighbors (K-최근접 이웃)
    • Distance Metrics (거리 척도)
    • Curse of Dimensionality (차원의 저주)
    • Weighted KNN (가중 KNN)
  • 1111-11-11, Support Vector Machines (서포트 벡터 머신) ⭐
    • Linear SVM (선형 SVM)
    • Kernel Trick (커널 트릭)
    • Soft Margin (소프트 마진)
    • Multi-class SVM (다중클래스 SVM)
  • 1111-11-11, Decision Trees (의사결정 나무) ⭐
    • Splitting Criteria (분할 기준)
      • Gini Impurity (지니 불순도)
      • Information Gain (정보 이득)
      • Gain Ratio (이득 비율)
    • Pruning (가지치기)
    • CART Algorithm (CART 알고리즘)

4.3 Ensemble Methods (앙상블 방법)

• Ensemble Learning (앙상블 학습) ⭐
  • 1111-11-11, Bagging Methods (배깅 방법)
    • Random Forest (랜덤 포레스트) ⭐
      • Bootstrap Aggregating
      • Feature Randomness (특성 무작위성)
      • Out-of-Bag Error (OOB 오차)
      • Feature Importance (특성 중요도)
    • Extra Trees (극단적 무작위 나무)
  • 1111-11-11, Boosting Methods (부스팅 방법) ⭐
    • AdaBoost (적응적 부스팅)
      • Weak Learner Weighting (약학습기 가중치)
      • Sample Reweighting (표본 재가중)
    • Gradient Boosting (경사 부스팅) ⭐
      • Gradient Boosting Machines (GBM)
      • Loss Function Optimization (손실함수 최적화)
      • Learning Rate and Trees (학습률과 나무 수)
    • XGBoost ⭐
      • Regularization Terms (정규화 항)
      • System Optimization (시스템 최적화)
      • Handling Missing Values (결측치 처리)
    • LightGBM ⭐
      • Histogram-based Algorithm (히스토그램 기반)
      • Leaf-wise Growth (리프 단위 성장)
      • Categorical Feature Support (범주형 특성 지원)
    • CatBoost
      • Ordered Boosting (순서 부스팅)
      • Native Categorical Handling (네이티브 범주형 처리)
  • 1111-11-11, Stacking (스태킹)
    • Meta-learner (메타학습기)
    • Blending (블렌딩)
  • 1111-11-11, Voting Classifiers (투표 분류기)
    • Hard Voting (하드 투표)
    • Soft Voting (소프트 투표)

5 Unsupervised Learning

5.1 Clustering (군집화)

• Clustering Methods (군집화 방법)
  • 1111-11-11, K-Means Clustering (K-평균 군집화) ⭐
    • Lloyd’s Algorithm (로이드 알고리즘)
    • Elbow Method (엘보우 방법)
    • K-Means++ Initialization (K-평균++ 초기화)
    • Mini-Batch K-Means (미니배치 K-평균)
  • 1111-11-11, Hierarchical Clustering (계층적 군집화)
    • Agglomerative Clustering (병합 군집화)
    • Divisive Clustering (분할 군집화)
    • Linkage Methods (연결 방법)
    • Dendrogram (덴드로그램)
  • 1111-11-11, DBSCAN (밀도 기반 군집화)
    • Density-Based Clustering (밀도 기반)
    • Core Points and Border Points (핵심점과 경계점)
    • Handling Noise (잡음 처리)
  • 1111-11-11, Gaussian Mixture Models (가우시안 혼합 모형)
    • Expectation-Maximization (EM 알고리즘)
    • Soft Clustering (소프트 군집화)
    • Model Selection (모형 선택)
  • 1111-11-11, Other Clustering Methods (기타 군집화 방법)
    • OPTICS
    • Mean Shift
    • Spectral Clustering (스펙트럼 군집화)

5.2 Dimensionality Reduction (차원 축소)

• Dimensionality Reduction Methods (차원 축소 방법)
  • 1111-11-11, Principal Component Analysis (주성분 분석) ⭐
    • Eigenvalue Decomposition (고유값 분해)
    • Variance Explained (설명된 분산)
    • Scree Plot (스크리 플롯)
    • Kernel PCA (커널 PCA)
  • 1111-11-11, Linear Discriminant Analysis (선형판별분석)
    • Fisher’s Linear Discriminant (피셔의 선형판별)
    • Between-class vs. Within-class Variance
  • 1111-11-11, t-SNE (t-분포 확률적 임베딩)
    • Perplexity Parameter (복잡도 매개변수)
    • KL Divergence Optimization (KL 발산 최적화)
    • Visualization Considerations (시각화 고려사항)
  • 1111-11-11, UMAP (균일 다양체 근사)
    • Topological Data Analysis (위상 데이터 분석)
    • Comparison with t-SNE
  • 1111-11-11, Autoencoders (오토인코더)
    • Vanilla Autoencoder
    • Denoising Autoencoder (잡음제거 오토인코더)
    • Variational Autoencoder (VAE)

5.3 Anomaly Detection (이상 탐지)

• Anomaly Detection Methods (이상 탐지 방법)
  • 1111-11-11, Statistical Methods (통계적 방법)
    • Z-score Method
    • Interquartile Range (IQR)
    • Mahalanobis Distance (마할라노비스 거리)
  • 1111-11-11, Isolation Forest (고립 숲)
    • Random Partitioning (무작위 분할)
    • Anomaly Score (이상 점수)
  • 1111-11-11, One-Class SVM (단일클래스 SVM)
  • 1111-11-11, Local Outlier Factor (LOF)
    • Local Density Estimation (지역 밀도 추정)

5.4 Association Rule Learning (연관 규칙 학습)

• Association Rules (연관 규칙)
  • 1111-11-11, Apriori Algorithm (Apriori 알고리즘)
    • Support, Confidence, Lift (지지도, 신뢰도, 향상도)
    • Frequent Itemsets (빈발 항목집합)
  • 1111-11-11, FP-Growth
  • 1111-11-11, Eclat Algorithm

6 Model Evaluation and Selection

6.1 Performance Metrics (성능 지표)

• Evaluation Metrics (평가 지표) ⭐
  • 1111-11-11, Classification Metrics (분류 지표)
    • Confusion Matrix (혼동 행렬)
    • Accuracy, Precision, Recall (정확도, 정밀도, 재현율)
    • F1-Score and F-beta Score
    • ROC Curve and AUC (ROC 곡선과 AUC)
    • Precision-Recall Curve (정밀도-재현율 곡선)
    • Matthews Correlation Coefficient (MCC)
    • Cohen’s Kappa (코헨의 카파)
  • 1111-11-11, Regression Metrics (회귀 지표)
    • Mean Squared Error (MSE, 평균제곱오차)
    • Root Mean Squared Error (RMSE)
    • Mean Absolute Error (MAE, 평균절대오차)
    • R-squared and Adjusted R-squared
    • Mean Absolute Percentage Error (MAPE)
  • 1111-11-11, Ranking Metrics (순위 지표)
    • Mean Average Precision (MAP)
    • Normalized Discounted Cumulative Gain (NDCG)
  • 1111-11-11, Clustering Metrics (군집화 지표)
    • Silhouette Score (실루엣 점수)
    • Davies-Bouldin Index
    • Calinski-Harabasz Index

6.2 Model Validation (모델 검증)

• Validation Strategies (검증 전략) ⭐
  • 1111-11-11, Train-Test Split (학습-테스트 분할)
  • 1111-11-11, Cross-Validation (교차 검증) ⭐
    • K-Fold Cross-Validation (K-겹 교차검증)
    • Stratified K-Fold (층화 K-겹)
    • Leave-One-Out (LOOCV)
    • Time Series Split (시계열 분할)
  • 1111-11-11, Bootstrap Methods (부트스트랩 방법)
    • Bootstrap Sampling (부트스트랩 샘플링)
    • Out-of-Bag Validation
    • .632 Bootstrap

6.3 Hyperparameter Tuning (하이퍼파라미터 튜닝)

• Hyperparameter Optimization (하이퍼파라미터 최적화) ⭐
  • 1111-11-11, Grid Search (그리드 탐색)
  • 1111-11-11, Random Search (무작위 탐색)
  • 1111-11-11, Bayesian Optimization (베이지안 최적화)
    • Gaussian Process (가우시안 과정)
    • Acquisition Functions (획득 함수)
  • 1111-11-11, Hyperband and ASHA
  • 1111-11-11, Automated Machine Learning (AutoML)
    • Auto-sklearn
    • TPOT
    • H2O AutoML

6.4 Model Selection (모델 선택)

• Model Selection Strategies (모델 선택 전략)
  • 1111-11-11, Information Criteria (정보 기준)
    • AIC (Akaike Information Criterion)
    • BIC (Bayesian Information Criterion)
    • MDL (Minimum Description Length)
  • 1111-11-11, Model Comparison (모델 비교)
    • Statistical Tests for Model Comparison
    • McNemar’s Test
    • Paired t-test
  • 1111-11-11, Baseline Models (베이스라인 모델)

7 Advanced Topics

7.1 Imbalanced Learning (불균형 학습)

• Handling Imbalanced Data (불균형 데이터 처리) ⭐
  • 1111-11-11, Resampling Techniques (리샘플링 기법)
    • Oversampling (과대표집)
      • Random Oversampling
      • SMOTE (Synthetic Minority Over-sampling)
      • ADASYN
    • Undersampling (과소표집)
      • Random Undersampling
      • Tomek Links
      • NearMiss
    • Combined Methods (결합 방법)
  • 1111-11-11, Cost-Sensitive Learning (비용 민감 학습)
    • Class Weights (클래스 가중치)
    • Cost Matrix (비용 행렬)
  • 1111-11-11, Ensemble Methods for Imbalanced Data
    • Balanced Random Forest
    • EasyEnsemble
    • RUSBoost

7.2 Probabilistic Models (확률 모형)

• Probabilistic Machine Learning (확률적 머신러닝)
  • 1111-11-11, Bayesian Methods (베이지안 방법)
    • Bayesian Linear Regression
    • Bayesian Logistic Regression
    • Prior Selection (사전분포 선택)
  • 1111-11-11, Gaussian Processes (가우시안 과정) ⭐
    • Kernel Functions (커널 함수)
    • Predictive Distribution (예측 분포)
    • Uncertainty Quantification (불확실성 정량화)
  • 1111-11-11, Hidden Markov Models (은닉 마르코프 모형)
    • Forward-Backward Algorithm
    • Viterbi Algorithm
  • 1111-11-11, Probabilistic Graphical Models (확률적 그래프 모형)
    • Bayesian Networks (베이지안 네트워크)
    • Markov Random Fields (마르코프 무작위장)

7.3 Online Learning (온라인 학습)

• Online and Incremental Learning (온라인 및 증분 학습)
  • 1111-11-11, Stochastic Gradient Descent (확률적 경사하강)
  • 1111-11-11, Online Learning Algorithms (온라인 학습 알고리즘)
    • Perceptron
    • Passive-Aggressive Algorithms
    • Online Gradient Descent
  • 1111-11-11, Concept Drift Detection (개념 표류 탐지)
    • ADWIN
    • DDM (Drift Detection Method)
  • 1111-11-11, Streaming Algorithms (스트리밍 알고리즘)

7.4 Semi-Supervised Learning (준지도학습)

• Semi-Supervised Methods (준지도 방법)
  • 1111-11-11, Self-Training (자기학습)
  • 1111-11-11, Co-Training (공동학습)
  • 1111-11-11, Label Propagation (레이블 전파)
  • 1111-11-11, Pseudo-Labeling (의사 레이블링)

7.5 Transfer Learning (전이학습)

• Transfer Learning Methods (전이학습 방법)
  • 1111-11-11, Domain Adaptation (도메인 적응)
  • 1111-11-11, Fine-tuning (미세조정)
  • 1111-11-11, Multi-task Learning (다중작업 학습)

7.6 Interpretability and Explainability (해석가능성)

• Model Interpretability (모델 해석가능성) ⭐
  • 1111-11-11, Feature Importance (특성 중요도)
    • Permutation Importance (순열 중요도)
    • Drop-Column Importance
  • 1111-11-11, Partial Dependence Plots (부분 의존 플롯)
  • 1111-11-11, SHAP (SHapley Additive exPlanations) ⭐
    • Shapley Values (섀플리 값)
    • TreeSHAP
    • KernelSHAP
  • 1111-11-11, LIME (Local Interpretable Model-agnostic Explanations)
  • 1111-11-11, Anchors and Counterfactuals (앵커와 반사실)

7.7 Fairness and Bias (공정성과 편향)

• Fairness in Machine Learning (머신러닝의 공정성)
  • 1111-11-11, Bias Detection (편향 탐지)
    • Demographic Parity (인구통계적 동등성)
    • Equal Opportunity (동등한 기회)
    • Disparate Impact (차별적 영향)
  • 1111-11-11, Bias Mitigation (편향 완화)
    • Pre-processing Methods (전처리 방법)
    • In-processing Methods (학습 중 방법)
    • Post-processing Methods (후처리 방법)
  • 1111-11-11, Fairness Metrics (공정성 지표)

8 ML Engineering

8.1 Production ML (프로덕션 ML)

• ML in Production (프로덕션 머신러닝)
  • 1111-11-11, Model Serialization (모델 직렬화)
    • Pickle and Joblib
    • ONNX
    • PMML
  • 1111-11-11, Model Serving (모델 서빙)
    • REST API
    • gRPC
    • Batch Prediction (배치 예측)
  • 1111-11-11, Model Monitoring (모델 모니터링) ⭐
    • Performance Monitoring (성능 모니터링)
    • Data Drift Detection (데이터 표류 탐지)
    • Model Drift Detection (모델 표류 탐지)
    • Feature Drift (특성 표류)
  • 1111-11-11, Model Versioning (모델 버전관리)
  • 1111-11-11, A/B Testing for Models (모델 A/B 테스트)

8.2 Scalability (확장성)

• Scaling Machine Learning (머신러닝 확장)
  • 1111-11-11, Distributed Training (분산 학습)
    • Data Parallelism (데이터 병렬화)
    • Model Parallelism (모델 병렬화)
  • 1111-11-11, Large-scale ML Frameworks (대규모 ML 프레임워크)
    • Apache Spark MLlib
    • Dask-ML
    • Ray
  • 1111-11-11, Feature Stores (특성 저장소)
  • 1111-11-11, Model Compression (모델 압축)
    • Quantization (양자화)
    • Pruning (가지치기)
    • Knowledge Distillation (지식 증류)

9 Special Applications

9.1 ML for Longitudinal Data (종단 데이터 ML)

• 2026-03-07, ML Overview: RSF, XGBoost, HMM, Lasso
• 2026-03-08, Random Survival Forest (RSF)
• 2026-03-08, XGBoost + 시간 피처 공학 (Temporal Feature Engineering)
• 2026-03-08, Hidden Markov Model (HMM) for Longitudinal Data
• 2026-03-08, Lasso / Elastic Net / glmmLasso — 종단 데이터의 변수 선택
• 통계적 기초 (LMM, GLMM, GEE): Statistics — Mixed Models & Longitudinal Analysis
• 딥러닝/강화학습 접근: Deep Learning — DL/RL for Longitudinal Data

9.2 Time Series Analysis (시계열 분석)

• Time Series Methods (시계열 방법)
  • 1111-11-11, Classical Methods (고전적 방법)
    • ARIMA Models
    • Exponential Smoothing (지수 평활)
    • Seasonal Decomposition (계절 분해)
  • 1111-11-11, ML for Time Series (시계열 ML)
    • Feature Engineering for Time Series
    • Lagged Features (지연 특성)
    • Rolling Statistics (이동 통계량)
  • 1111-11-11, Forecasting (예측)
    • Prophet
    • LightGBM for Time Series
    • XGBoost for Time Series

9.3 Natural Language Processing (자연어처리)

• 2026-03-27, Fine-tuning 학습 데이터 샘플 수 추정 — KoBERT 14-class 분류 사례
• 2026-04-15, Train/Val/Test 분할 비율 선택 — 데이터 규모별 실무 가이드
• Classical NLP with ML (머신러닝 기반 NLP)
  • 1111-11-11, Text Preprocessing (텍스트 전처리)
    • Tokenization (토큰화)
    • Stemming and Lemmatization (어간추출과 표제어추출)
    • Stop Words Removal (불용어 제거)
  • 1111-11-11, Text Representation (텍스트 표현)
    • Bag of Words (단어 가방)
    • TF-IDF
    • Word Embeddings (단어 임베딩)
      • Word2Vec
      • GloVe
      • FastText
  • 1111-11-11, Text Classification (텍스트 분류)
    • Naive Bayes for Text
    • SVM for Text
  • 1111-11-11, Topic Modeling (토픽 모델링)
    • Latent Dirichlet Allocation (LDA)
    • Non-negative Matrix Factorization (NMF)

9.4 Computer Vision (컴퓨터 비전)

• Classical Computer Vision with ML (머신러닝 기반 컴퓨터 비전)
  • 1111-11-11, Image Features (이미지 특성)
    • HOG (Histogram of Oriented Gradients)
    • SIFT (Scale-Invariant Feature Transform)
    • SURF
  • 1111-11-11, Image Classification (이미지 분류)
    • SVM for Images
    • Random Forest for Images
  • 1111-11-11, Object Detection Basics (객체 탐지 기초)
    • Sliding Window (슬라이딩 윈도우)
    • Image Pyramids (이미지 피라미드)

9.5 Recommender Systems (추천 시스템)

• Recommendation Algorithms (추천 알고리즘)
  • 1111-11-11, Collaborative Filtering (협업 필터링)
    • User-based CF (사용자 기반)
    • Item-based CF (아이템 기반)
    • Matrix Factorization (행렬 분해)
      • SVD (특이값 분해)
      • ALS (교대 최소자승)
  • 1111-11-11, Content-Based Filtering (내용 기반 필터링)
  • 1111-11-11, Hybrid Methods (하이브리드 방법)
  • 1111-11-11, Evaluation Metrics (평가 지표)
    • Precision@K and Recall@K
    • MAP and NDCG
    • Coverage and Diversity

10 Key Resources

10.1 Books

• Foundations:
  • Bishop (2006). “Pattern Recognition and Machine Learning”
  • Murphy (2012). “Machine Learning: A Probabilistic Perspective”
  • Hastie, Tibshirani, Friedman (2009). “The Elements of Statistical Learning” ⭐
• Practical:
  • Géron (2019). “Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow”
  • Kuhn and Johnson (2013). “Applied Predictive Modeling”
• Advanced:
  • Raschka and Mirjalili (2019). “Python Machine Learning”
  • Chollet (2021). “Deep Learning with Python”

10.2 Online Courses

• Andrew Ng - Machine Learning (Coursera) ⭐
• Fast.ai - Practical Deep Learning for Coders
• Stanford CS229 - Machine Learning

10.3 Papers

• Random Forests: Breiman (2001). “Random Forests”
• XGBoost: Chen and Guestrin (2016). “XGBoost: A Scalable Tree Boosting System”
• SHAP: Lundberg and Lee (2017). “A Unified Approach to Interpreting Model Predictions”