Machine learning notes

The following notes I started writing from 2020. I meant for them to be a more comprehensive offline study object for the students, for whom I could never quite find an appropriate machine learning textbook for a beginner.  Since then, many great ML textbooks that are more mathematically inclined have appeared. Still, I use this version because I'm most familiar with it now. But it is by no means purely unique or comprehensive.

I do apologize if some elements of the notes are not sophisticatedly cited -- this I am still working on. For now, the vague list of acknowledgements are

• The previous instructor of this course at SBU, of whom the structure and topic overviews are inherited from

• All my past instructors, of whom I'm sure I borrowed some examples and ways of explaining from, but they are so internalized now that I have no idea who told me what.

• My current students, who if they suggest good improvements are given extra credit points (which they have done fervently and with very high quality)

Citations: I would actually prefer it if you didn't cite this. There's nothing new here, it's just my take on existing subjects. As I start curating the citations for each chapter, you can just use the ones in the chapters themselves.  But if you absolutely felt the need, please use this bibtex:

@misc{sun2023mlnotes,
author = {Yifan Sun},
title = {Machine Learning Notes},
year = {2023},
url = {https://sites.google.com/site/yifansunwebsite/teaching-materials/machine-learning-course-notes}} 

• Probability review

• Point estimation

• Statistical decision theory

• Probabilistic modeling

• Linear algebra review

• Linear regression

• Unconstrained optimization

• Constrained optimization

• Logistic regression

• SVM

• Dual SVM

• Multiclass SVM

• Expectation maximization

• Ensembling

• Generalization

• Representational learning