I just start to learn Machine Learning, and follow the OCW below, you can see the material on youtube, couresa, or his page
https://www.youtube.com/channel/UC9Wi1Ias8t4u1OosYnHhi0Q https://www.youtube.com/channel/UC2ggjtuuWvxrHHHiaDH1dlQ https://www.youtube.com/channel/UCBtDVX6tpl1SCPsT5SQEnYQ
In machine learning, there are four types of learning, supervised learning, un-supervised learning, semi-supervised learning, and reinforcement learning. Let's introduce them.
- Supervised learning
Supervised learning needs the data with label, and use the labeled data to train a model, then input a data without label and predict its label. So it needs much time to label much data. This algorithm is for classification in common use. Here are some algorithms in this folder is this type: SVM, regression, decision tree, k-nearest neiborhood, Nueral network.
- Un-supervised learning
Un-supervised learning dosen't need labeled-data, but the error may be big because of the noise. This algorithm is for cluster in common use. Here are some algorithms in this folder is this type: K-means
- Semi-supervised learning
Semi-supervised learning means some data with label and some are not, you can write an algorithm to help you label the data without labeled after classifying, so it is in common practice.
- Reinforcement learning
Reinforcement learning is like the AI in some movie or anime, the algorithm will learn the behavior by itself based on the score, so it can correct itself. The famous algorithm in this type is Deep learning.
In this folder, it is recommanded follow the order
- Percetron Learning Algorithm.pdf https://github.com/JrPhy/MachineLearning/blob/master/Perceptron%20Learning%20Algorithm/Percetron%20Learning%20Algorithm.pdf
- Support Vector Machine.pdf https://github.com/JrPhy/MachineLearning/blob/master/Support%20Vector%20Machine/Support%20Vector%20Machine.pdf
- Nonlinear transform.pdf https://github.com/JrPhy/MachineLearning/blob/master/Support%20Vector%20Machine/Nonlinear%20transform.pdf
- SVM kernel trick and soft-margin SVM.pdf https://github.com/JrPhy/MachineLearning/blob/master/Support%20Vector%20Machine/SVM%20kernel%20trick%20and%20soft-margin%20SVM.pdf
- Regularization.pdf https://github.com/JrPhy/MachineLearning/blob/master/Support%20Vector%20Machine/Regularization.pdf
- Kernel Logistic Regression.pdf https://github.com/JrPhy/MachineLearning/blob/master/Support%20Vector%20Machine/Kernel%20Logistic%20Regression.pdf
- Support Vector Regression.pdf https://github.com/JrPhy/MachineLearning/blob/master/Support%20Vector%20Machine/Support%20Vector%20Regression.pdf
- Blending and Bagging.pdf https://github.com/JrPhy/MachineLearning/blob/master/Bootstrap/Blending%20and%20Bagging.pdf
- Decision tree and random forest.pdf https://github.com/JrPhy/MachineLearning/blob/master/Bootstrap/Decision%20tree%20and%20random%20forest.pdf
- Gradient descent.pdf https://github.com/JrPhy/MachineLearning/blob/master/Bootstrap/Gradient%20descent.pdf
- Gradient Boosted Decision Tree.pdf https://github.com/JrPhy/MachineLearning/blob/master/Bootstrap/Gradient%20Boosted%20Decision%20Tree.pdf