机器学习 (Machine learning) 美本/硕 要点概览

简单归纳一下 美本/硕 机器学习课程的要点

Category I: Supervised learning (监督学习)

I.A Regression (回归)
1. Linear regression (线性回归)
2. Linear regression with nonlinear basis (非线性基线性回归)
3. Neural network (神经网络，可通用于回归/分类)

I.B Classification (分类)
1. K-nearest neighbor classifier (K-近邻分类)
2. Perceptron algorithm (感知器算法)
3. Logistic regression (对数几率回归)
4. Multiclass classification (多元分类)
5. Convolutional neural network (卷积神经网络，多用于图像分类）
6. Support vector machine (支持向量机)
7. Decision tree (决策树)
8. Random Forest (随机森林)
9. Boosting (提升方法)

I.C Supervised learning tricks
1. Overfitting/Underfitting (过度/欠拟合)
2. Kernel method (内核方法)
3. Lagrangian duality (拉格朗日对偶性)

Category II: Unsupervised learning (无监督学习)

1. K-means clustering (K-平均聚类)
2. Gaussian mixture model (高斯混合模型)
3. Expectation-maximization (EM) algorithm (期望最大化算法)
4. Maximum likelihood estimation (最大似然估计)
5. (Kernel) density estimation ("核"密度估计)
6. Naive Bayes classifier (朴素贝叶斯分类器)
7. Principal component analysis (主成分分析)
8. Hidden Markov model (隐马尔可夫模型)

Category III: Reinforcement learning (强化学习)...

希望上述归纳有用！

EM算法是无监督学习吗？

槽点略多的总结