This is my solutions for course CS229 - Machine Learning of Stanford University, taught by Andrew Ng. The lecture notes are original versions from the main page of Stanford.
- cs229-notes9 is lecture about factor analysis
- main_notes is lecture note for all the course. However, since this course is taught from 2017 which might be lack of some important concept nowadays (variational autoencoder, ICA,...), I recommend some other sources such as CS229 summer 2019 or MIT textbooks.
- The course get through all fundamentals of supervised learning containing Linear Regression, Logistic Regression, the theory of Gaussian Distribution Function and some hands-on lab such as Quatar Regression
- Problem set and its solution: PS1
- This section dives more into the learning theory of supervised learning such as model calibration, training stability
- More complex learning models are introduced such as Support Vector Machine, Naive Bayes classifier and their applications
- Problem set and its solution: PS2
- Unsupervised Learning is a space of learning without label in which let the models find the hidden patterns in data themselves. Some popular models in this research aspect are introduced in this course such as Neural Networks, K-Means, EM algorithm
- After this section, you can also be able to understand more about probabilistic aspect in machine learning world
- Problem set and its solution: PS3
- Those 3 aspects in machine learning world nowadays are the most popular model and research aspects. The final section give you some fundamentals about Reinforcement Learning such as Markov decision process, Value iteration, etc.
- You will have a chance to implement application of RL in inverted pendulum, DL in image classification and EM algorithm mathematical base.
- Problem set and its solution: PS4