2012MATLAB校园巡回研讨会.PDF

  1. 1、本文档共31页,可阅读全部内容。
  2. 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. 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. 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)。
查看更多
2012MATLAB校园巡回研讨会

Copyright ? 2012 by TeraSoft, Inc. 2012 MATLAB校園巡迴研討會 統計運算技巧/資料探勘之應用 Agenda ? 13:30~13:50 R2011b與R2012a新功能介紹 ? 13:50~14:50 利用 MATLAB 產品進行運算統計 ? 14:50~15:10 休息時間 Copyright ? 2012 by TeraSoft, Inc. ? 15:10~16:10 MATLAB 於資料探勘之應用 ? 16:10~16:30 問題討論 What’s New for MATLAB with R2011b ? MATLAB ? New Spreadsheet Import Tool ? Load and Save Parts of Variables ? Automatic renaming all variables ? Statistics Copyright ? 2012 by TeraSoft, Inc. ? Ridge regression ? Lasso ? Elastic net What’s New for MATLAB with R2011b (Cont.) ? GPU Computing ? GPU-enabled functions can be deployed using MATLAB Compiler Copyright ? 2012 by TeraSoft, Inc. ? New Product: ? MATLAB Coder What’s New for MATLAB with R2012a ? Statistics Toolbox ? Regression analysis is simplified ? Fitting, prediction, and plotting ? New content to highlight the new features Copyright ? 2012 by TeraSoft, Inc. ? MATLAB Virtual Conference – March 28, 2012 ? New Capabilities for Regression and Curve Fitting ? Public webinar – April 17, 2012 ? Regression Analysis with MATLAB: New Statistics Toolbox Capabilities in R2012a Agenda ? 13:30~13:50 R2011b與R2012a新功能介紹 ? 13:50~14:50 利用 MATLAB 產品進行運算統計 ? 14:50~15:10 休息時間 Copyright ? 2012 by TeraSoft, Inc. ? 15:10~16:10 MATLAB 於資料探勘之應用 ? 16:10~16:30 問題討論 Section Outline ? Feature Selection, Regularization, and Shrinkage with MATLAB ? Feature selection techniques ? All possible subsets Copyright ? 2012 by TeraSoft, Inc. ? Sequential feature selection ? Demo ? Regularization and shrinkage ? Lasso, Ridge Regression, Elastic Net ? Demo Subset Selection and Coefficient Shrinkage Why do we care?  Design Considerations – Reduce Memory Copyright ? 2012 by TeraSoft, Inc. – Improve Speed  Interpretability – Simple models are easier to interpret  Predictive Accuracy (Overfitting) – How well can we fit this particular data set? – How well can the results be generalized to other data sets? Predictive Accuracy Copyright ? 2012 by TeraSoft, Inc. Cross Validatio

文档评论(0)

***** + 关注
实名认证
内容提供者

该用户很懒,什么也没介绍

版权声明书
用户编号:8016031115000003

1亿VIP精品文档

相关文档