Lecture Notes - Statistical Learning Theory

• Syllabus
• part 1: PAC Learnability
• part 2: Learnability of Infinite Hypothesis Classes
• part 3: Bias-Variance Tradeoff and the No-Free-Lunch Theorem
• part 4: Non-Uniform Learnability
• part 5: Advanced Theory
• part 6: Boosting
• part 7: Universal Approximation of Neural Networks
• part 8: Generalized Linear Models
• part 9: Stability, Regularization, and Online Classification
• part 10: Modern Theory - Generalization in the Overparameterized Regime