In 2017 Fall CS534, We learned many machine learning algorithms.
In Homework 1, we mainly dealt with linear regression and classification, like SGD, naive Bayes, Linear regression and ridge/lasso regression. One major step is data processing, how to convert categorical data to numerical data, how to deal with missing values. RMSE and AUC are also used in the evaluation step
In HW2, first question is about Bias-Variance trade off. Then,we implemented LDA, QDA, BLB and logistic regression.
In HW3, We implemented Gradient Boosting and Decision Tree, as well as Regression Tree. How to determinate the optimal parameter is very important, grid search is applied in this step. Also, we learned how to do validation to identify the best model.
In HW4, we first learned about the idea of kernel. Then we implemented percenpton, random forest and SVM.
In HW5, we implemented NN.