COMP5318 Knowledge Discovery and Data Mining_2011 Semester 1_week7EM_dimred.pdf

下载文档

8
0
约 12页
2017-04-11 发布于江苏
举报
版权申诉
保障服务

COMP5318 Knowledge Discovery and Data Mining_2011 Semester 1_week7EM_dimred.pdf

1、本文档共12页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。

COMP5318 Knowledge Discovery and Data Mining_2011 Semester 1_week7EM_dimred

Topics for the day ? Expectation Maximisation II ? Dimensionality Reduction ? PCA ? MDS ? Isomap 1 Announcements ? Marks for Assignment 1 will be available in 2 weeks. ? Assignment 2 will be available in web this week. 2 Expectation Maximisation (1) 3 ? Probabilistic method for soft clustering. ? Direct method that assumes k clusters:{c1, c2,… ck} ? Soft version of k-means. ? Assumes a probabilistic model of categories that allows computing p(z | x) for each category, z, for a given example, x. !#$%(()*$+)(,-)./0* 4 %(()*$1-2#,#-$3(/2/0* Likelihood function 5 45)22$67089$7):#;)00 6 7 !#$%()*+,(-(.()* ? EM is an algorithm for finding MLE or MAP for problems with hidden variables ? Key intuition: if we knew what cluster each point belonged to (i.e., the z_i variables), we could partition the data and find the MLE for each cluster separately ? E step: infer responsibility of each cluster for each data point ? M step: optimize parameters using “filled in” data z ? Repeat until convergence !! ! #! ! $!!%# Expectation Maximisation (2) ? Gaussian mixture model ? Mixing coefficients ? Maximise the log-likelihood 8 9.2. Mixtures of Gaussians 433 (a) 0 0.5 1 0 0.5 1 (b) 0 0.5 1 0 0.5 1 (c) 0 0.5 1 0 0.5 1 Figure 9.5 Example of 500 points drawn from the mixture of 3 Gaussians shown in Figure 2.23. (a) Samples from the joint distribution p(z)p(x|z) in which the three states of z, corresponding to the three components of the mixture, are depicted in red, green, and blue, and (b) the corresponding samples from the marginal distribution p(x), which is obtained by simply ignoring the values of z and just plotting the x values. The data set in (a) is said to be complete, whereas that in (b) is incomplete. (c) The same samples in which the colours represent the value of the responsibilities γ(znk) associated with data point xn, obtained by plotting the corresponding point using proportions of red, blue, and green ink given by γ(znk) for k = 1, 2, 3, respectively matrix X in w