32513 31005 - Summary

These notes follow the first five modules of Advanced Data Analytics Algorithms, Machine Learning. They move from the definition of a learning problem to the criterion, model family, and algorithm used to construct a working predictor.

Module	Main concepts	Notes
1. What Is Machine Learning?	Learning from examples, generalization, and when machine learning is useful	Read Module 1
2. Statistical Learning Framework	Data distributions, hypotheses, criteria, and learning algorithms	Read Module 2
3. Learning Criterion and Loss Functions	Regression and classification losses, information theory, structured outputs, and empirical risk	Read Module 3
4. Hypothesis Design and Unsupervised Models	Basis functions, neural networks, kernels, representation learning, and generative models	Read Module 4
5. Training Algorithms and Gradient Optimization	Closed-form and data-based models, gradient descent, and adaptive optimizers	Read Module 5

The module pages contain the detailed explanations, equations, examples, and comparisons. Later modules can be added to this course as their notes are prepared.