ÿþ<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns="http://www.w3.org/TR/REC-html40"> <head> <meta http-equiv=Content-Type content="text/html; charset=unicode"> <meta name=ProgId content=Word.Document> <meta name=Generator content="Microsoft Word 10"> <meta name=Originator content="Microsoft Word 10"> <link rel=File-List href="June30-2003_files/filelist.xml"> <link rel=Edit-Time-Data href="June30-2003_files/editdata.mso"> <!--[if !mso]> <style> v\:* {behavior:url(#default#VML);} o\:* {behavior:url(#default#VML);} w\:* {behavior:url(#default#VML);} .shape {behavior:url(#default#VML);} </style> <![endif]--> <title>Seminar: July 09, 2002</title> <!--[if gte mso 9]><xml> <o:DocumentProperties> <o:Author>Dide Ergüven</o:Author> <o:LastAuthor>Dide Ergüven</o:LastAuthor> <o:Revision>3</o:Revision> <o:TotalTime>1</o:TotalTime> <o:Created>2003-05-20T13:13:00Z</o:Created> <o:LastSaved>2003-06-26T13:09:00Z</o:LastSaved> <o:Pages>1</o:Pages> <o:Words>440</o:Words> <o:Characters>2509</o:Characters> <o:Company>Bilkent Üniversitesi</o:Company> <o:Lines>20</o:Lines> <o:Paragraphs>5</o:Paragraphs> <o:CharactersWithSpaces>2944</o:CharactersWithSpaces> <o:Version>10.2625</o:Version> </o:DocumentProperties> </xml><![endif]--><!--[if gte mso 9]><xml> <w:WordDocument> <w:Zoom>120</w:Zoom> <w:SpellingState>Clean</w:SpellingState> <w:GrammarState>Clean</w:GrammarState> <w:BrowserLevel>MicrosoftInternetExplorer4</w:BrowserLevel> </w:WordDocument> </xml><![endif]--> <style> <!--a:link {font-weight: Regular;} a:visited {font-weight: Regular;} a:active {font-weight: Regular;} A:hover { color: #A0A0A0 } h2 {FONT-FACE: Bold;} /* Font Definitions */ @font-face {font-family:Verdana; panose-1:2 11 6 4 3 5 4 4 2 4; mso-font-charset:162; mso-generic-font-family:swiss; mso-font-pitch:variable; mso-font-signature:536871559 0 0 0 415 0;} @font-face {font-family:"Century Gothic"; panose-1:2 11 5 2 2 2 2 2 2 4; mso-font-charset:162; mso-generic-font-family:swiss; mso-font-pitch:variable; mso-font-signature:647 0 0 0 159 0;} @font-face {font-family:"Trebuchet MS"; panose-1:2 11 6 3 2 2 2 2 2 4; mso-font-charset:162; mso-generic-font-family:swiss; mso-font-pitch:variable; mso-font-signature:647 0 0 0 159 0;} /* Style Definitions */ p.MsoNormal, li.MsoNormal, div.MsoNormal {mso-style-parent:""; margin:0cm; margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:12.0pt; font-family:"Times New Roman"; mso-fareast-font-family:"Times New Roman"; color:black;} h2 {mso-margin-top-alt:auto; margin-right:0cm; mso-margin-bottom-alt:auto; margin-left:0cm; mso-pagination:widow-orphan; mso-outline-level:2; font-size:12.0pt; font-family:Verdana; color:black; font-weight:bold;} h3 {mso-margin-top-alt:auto; margin-right:0cm; mso-margin-bottom-alt:auto; margin-left:0cm; mso-pagination:widow-orphan; mso-outline-level:3; font-size:13.5pt; font-family:"Times New Roman"; color:black; font-weight:bold;} a:link, span.MsoHyperlink {color:blue; mso-text-animation:none; text-decoration:none; text-underline:none; text-decoration:none; text-line-through:none;} a:visited, span.MsoHyperlinkFollowed {color:blue; mso-text-animation:none; text-decoration:none; text-underline:none; text-decoration:none; text-line-through:none;} p {mso-margin-top-alt:auto; margin-right:0cm; mso-margin-bottom-alt:auto; margin-left:0cm; mso-pagination:widow-orphan; font-size:12.0pt; font-family:"Times New Roman"; mso-fareast-font-family:"Times New Roman"; color:black;} span.SpellE {mso-style-name:""; mso-spl-e:yes;} span.GramE {mso-style-name:""; mso-gram-e:yes;} @page Section1 {size:612.0pt 792.0pt; margin:70.85pt 70.85pt 70.85pt 70.85pt; mso-header-margin:35.4pt; mso-footer-margin:35.4pt; mso-paper-source:0;} div.Section1 {page:Section1;} --> </style> <!--[if gte mso 10]> <style> /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin:0cm; mso-para-margin-bottom:.0001pt; mso-pagination:widow-orphan; font-size:10.0pt; font-family:"Times New Roman";} </style> <![endif]--><!--[if gte mso 9]><xml> <o:shapedefaults v:ext="edit" spidmax="3074"/> </xml><![endif]--><!--[if gte mso 9]><xml> <o:shapelayout v:ext="edit"> <o:idmap v:ext="edit" data="1"/> </o:shapelayout></xml><![endif]--> </head> <body bgcolor=white lang=EN-US link=blue vlink=blue style='tab-interval:36.0pt'> <div class=Section1> <p class=MsoNormal> <link rev=made href="mailto:guvenir@cs.bilkent.edu.tr"> <span style='font-family:Arial'><a href="http://www.bilkent.edu.tr" title="Bilkent Üniversitesi"><span style='font-size:13.5pt;color:navy'><!-- OWNER_INFO="Bilkent University, Department of Computer Engineering -->Bilkent University</span></a><br> <a href="http://www.cs.bilkent.edu.tr" title="Bilgisayar Mühendislii Bölümü"><span style='font-size:13.5pt;color:red'>Department of Computer Engineering</span></a> <o:p></o:p></span></p> <div class=MsoNormal align=center style='text-align:center'><span style='font-family:Arial'> <hr size=2 width="100%" noshade color=navy align=center> </span></div> <p class=MsoNormal align=center style='text-align:center'><b><span style='font-size:13.5pt;font-family:"Century Gothic";mso-bidi-font-family:Arial; color:#007744'>S E M I N A R</span></b><span style='font-family:"Century Gothic"; mso-bidi-font-family:Arial'> <o:p></o:p></span></p> <h2 align=center style='text-align:center'><span style='font-size:18.0pt; font-family:"Century Gothic";mso-bidi-font-family:Arial;color:navy;font-weight: normal'>ON-LINE Feature selection - the way to go </span><span style='mso-bidi-font-family:Arial'><o:p></o:p></span></h2> <p align=center style='text-align:center'><span style='font-family:"Century Gothic"; mso-bidi-font-family:Arial'>&nbsp;<o:p></o:p></span></p> <h3 align=center style='text-align:center'><span style='font-family:"Century Gothic"'>Prof. <span class=SpellE>Nikhil</span> R. Pal, Ph. D.<o:p></o:p></span></h3> <p align=center style='text-align:center'><span style='font-size:13.5pt; font-family:"Century Gothic";mso-bidi-font-family:Arial;color:gray'>Indian Statistical Institute <o:p></o:p></span></p> <p align=center style='text-align:center'><b><span style='font-size:10.0pt; font-family:"Century Gothic";mso-bidi-font-family:Arial;color:gray'>&nbsp;<o:p></o:p></span></b></p> <p align=center style='text-align:center'><span style='font-family:"Century Gothic"; mso-bidi-font-family:Arial'>&nbsp;<o:p></o:p></span></p> <p class=MsoNormal style='mso-margin-top-alt:auto;margin-bottom:12.0pt'><span style='font-family:"Trebuchet MS"'>Literature on feature selection deals with methods that are &quot;off-line&quot; in nature. Typically, some index is computed on each feature or on subsets of features. Then features are either ranked or some subset is selected. Either a few top ranked features or the selected subset is then used to design the system. Ranking is not a good philosophy, because the two top-ranked features are not necessarily the best two (<span class=SpellE><span class=GramE>eg</span></span><span class=GramE>.,</span> correlated features). It is worth noting here that quality of a feature depends on the TOOL being used and the PROBLEM being solved. So we introduce a novel concept of &quot;ON-LINE&quot; feature selection where the system picks up the required features along with training of the system - the idea is to associate a gate with each feature and keep the gate almost closed at the beginning of the training. The training process opens these gates (features), which are important and shuts the gates more tightly which are bad or redundant. In this context, we will explain three systems. The first system is designed for the multi-layer <span class=SpellE>perceptron</span> type networks. The system is applicable to both classification and function approximation type problems. The second system is built based a <span class=SpellE>neuro</span>-fuzzy framework, which uses a five layer network for solving function approximation type problems. Finally, this system is modified for dealing with classification type problems. The second part deals with sensor selection or group feature selection. (<span class=GramE>to</span> my knowledge this is also for the first time) For many applications, the input comes from different sensors. For example, in case of an intelligent weld inspection system, the sensors could be X-ray image, Acoustic emission, <span class=GramE>eddy</span> current and so on. The signal obtained from each sensor is used to compute several features. For example, the X-ray image can be used to compute several co-occurrence based features. In such cases, a more challenging problem comes - selection of sensors (in other words, selection of groups of features, where each group is computed using the signal obtained from a particular sensor). Clearly, if the number of necessary sensors can be reduced, the hardware cost of the system, the design complexity of the system and the cost (both in terms of time and money) of decision making can be drastically reduced. This problem, to our knowledge, has not been addressed in the literature. We will discuss two systems for online sensor selection. The first approach is applicable to multi-layer <span class=SpellE>perceptron</span> type networks while the second method is for radial basis function type network. </span></p> <p><span style='font-family:Verdana;mso-bidi-font-family:Arial'><br> &nbsp;</span><span style='font-family:Arial'><o:p></o:p></span></p> <p><b><span style='font-family:Arial;color:red'>DATE:</span></b><b><span style='font-family:Arial'> June 30, 2003, Monday @ 14:30 <br> </span></b><b><span style='font-family:Arial;color:red'>PLACE:</span></b><b><span style='font-family:Arial'> EA-409</span></b><span style='font-family:Arial'><o:p></o:p></span></p> </div> </body> </html>