教学目标：
本课程的教学目的是系统介绍机器学习的基础算法，如线性回归、感知机、逻辑斯蒂回归、Softmax回归、支持向量机、朴素贝叶斯模型等；讲述机器学习中的核心概念与核心思想，如模型假设、模型学习、模型推理、模型评估、模型选择、模型泛化等；使得学生掌握基础的机器学习理论与实践能力，能够运用常见机器学习模型解决分类、回归、聚类等任务。

课程内容：
1.  Introduction
1.1 What is machine learning
1.2 The History of machine learning
1.3 Real-world applications of machine learning
2.  Basic concepts in machine Learning
2.1 Supervised machine learning
2.2 Semi-supervised, unsupervised machine learning
2.3 Cost function and model learning
2.4 Evaluation and model selection
2.5 Over-fitting and Regularization
3.  Linear regression
3.1 Regression and classification
3.2 Least mean square criterion
3.3 Maximum likelihood estimation
3.4 Application of linear regression
4.  Logistic regression
4.1 Model assumptions
4.2 Model learning
4.3 Gradient descent, Newton’s method, and Quasi-Newton
4.4 Softmaxregresson
5.  Perceptron& Neural Networks
5.1 The perceptron criterion
5.2 Parameter learning
5.3 Stochastic gradient descent
5.4 Neural Networks
5.5Back-propagation algorithm
5.6 Recent advances in deep learning
6.  Support vector machines
6.1 Maximum margin criterion
6.2 Dualoptimzation
6.3 Soft-margin SVM
6.4 Kernel functions
6.5 How to use the LIBLINEAR toolkit
7. Na?ve Bayes
7.1 Generative model vs. Discriminative model
7.2 Multinomial event model
7.3 Parameter learning based on MLE
7.4 Bayes decision rule
7.5 Applications of Na?ve Bayes
8. Machine learning practice
8.1 Implementation of supervised machine learning algorithms
8.2 Design of a text classification system

适用学生：
全日制硕士    非全日制硕士    留学硕士    进修硕士    硕博连读    本科直博    全日制博士    留学博士    进修博士    在职专硕    其他    

预修课程：
高数、概率论、线性代数

参考书目：
Textbooks:
1.Tom M. Mitchell, Machine Learning, McGraw Hill, 1997
2.Bishop C. M. Pattern recognition and machine learning. New York: springer, 2006.
3.Stanford online course: http://cs229.stanford.edu/

备注：