机器学习与概率图模型中科院自动化所系列报告(王立威)25443.pptx

《机器学习与概率图模型中科院自动化所系列报告(王立威)25443.pptx》由会员分享，可在线阅读，更多相关《机器学习与概率图模型中科院自动化所系列报告(王立威)25443.pptx（122页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、Machine Learning and Graphical Models王立威北京大学信息科学技术学院(Lecture I)Outline A brief overview of Machine Learning Graphical Models Representation Inference Learning2 Definition of Machine Learning:Learning from experiences.“A computer program is said to learn from experience E with respect to some class o

2、f tasks T and performance measure P,if its performance at tasks in T,as measured by P,improves with experience E.”-Tom Mitchell3“Classical”Machine Learning Tasks:Classification:spam filter,face recognition,Regression Hooks law,Keplers law,Ranking Search engine Probability(Distribution)Estimation4“Cl

3、assical”Machine Learning Algorithms Classification SVM Boosting Random Forest Bagging(Deep)Neural Networks Regression Lasso Boosting5Support Vector Machines(SVMs)SVM:the large margin classifier SVM:hinge loss minimization+regularizationBoosting Boosting:(implicit)large margin classifier Boosting:exp loss minimization(+regularization)