第一问是要实现neural network, boosted decision trees, random forests. 比较他们的性能。
第二问是关于conditional probability的理论问题加编程问题，需要根据题目的公式以及dataset计算一个mean absolute error.
第三问是关于OLS（ordinary least square）的理论加计算，需要根据数据集以及书中的公式进行计算。
Problem 1 (Non-linear classiers; 20 points). For this problem, you will need to learn to use
software libraries for at least two of the following non-linear classier types:
• neural networks;
• boosted decision trees (i.e., boosting, with the \weak learner” being a decision tree learner);
• random forests.
All of these are available in scikit-learn and the MATLAB Statistics and Machine Learning toolbox,
although you may also use other external libraries (e.g., Tensor ow1 for neural networks, XGBoost2
for boosted decision trees). You are welcome to implement learning algorithms for these classiers
yourself, but this is neither required nor recommended.