# Register now to access 7 million high quality study materials (What's Course Hero?) Course Hero is the premier provider of high quality online educational resources. With millions of study documents, online tutors, digital flashcards and free courseware, Course Hero is helping students learn more efficiently and effectively. Whether you're interested in exploring new subjects or mastering key topics for your next exam, Course Hero has the tools you need to achieve your goals.

5 Pages

### Lecture8_handout

Course: FDA 2008, Fall 2009
School: Cornell
Rating:

Word Count: 804

#### Document Preview

BTRY 6150: Applied Functional &lt;a href=&quot;/keyword/data-analysis/&quot; &gt;data analysis&lt;/a&gt; : Constrained Smoothing BTRY 6150: Applied Functional &lt;a href=&quot;/keyword/data-analysis/&quot; &gt;data analysis&lt;/a&gt; : Constrained Smoothing Constrained Functions Positive Smoothing We know that angular acceleration must be positive: Text: Chapter 6...

Register Now

#### Unformatted Document Excerpt

Coursehero >> New York >> Cornell >> FDA 2008

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing Constrained Functions Positive Smoothing We know that angular acceleration must be positive: Text: Chapter 6 There are some situations in which we want to include known restrictions about x(t). x(t) is always positive x(t) is always increasing (or decreasing) x(t) is a density Idea: Enforce these conditions by transforming x(t). a(t) = [D 2 x(t)]2 + [D 2 y (t)]2 But negative values can occur because of smoothing/basis bias. BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing Positive Smoothing We want to ensure that x(t) &gt; 0. Observation: e w : ( , ) (0, ) So try the transformation x(t) = e W (t) with W (t) = (t)c and penalize the roughness of W (t) Estimating a Positive Smooth We now want to minimize n PENSSE (W ) = i=1 yi e W (ti ) 2 + [LW (t)]2 dt This does not have an explicit formula. But it is convex there is only one minimum. Requires numerical optimization, but this is generally fast. BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing Positive Smoothing Vancouver Precipitation Constraints imply smoothness (of a certain type) tend to need less smoothing on W . Monotone Smoothing Berkeley growth study heights aged 1 - 18 BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing Monotone Smoothing Estimating a Monotone Smooth We now want to minimize We need x(t) always increasing: n Dx(t) &gt; 0 suggests Dx(t) = e W (t) x(t) = + again, W (t) = (t)c t t0 PENSSE (W ) = i=1 yi ti e t0 W (s) 2 ds + [LW (t)]2 dt e W (s) ds No explicit formula No good formula for the integral Still a convex problem; numerics work fairly quickly Note, LW (t) = D 2 W (t) suggests that any x(t) = + e t is smooth. BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing Density Estimation Position of Beetles in Angles Density Estimation x(t) a density positive, integrates to 1 x(t) = e W (t) / But we observe only t1 , . . . , tn . Need to nd an objective to minimize. e W (t) dt BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing Penalized Likelihood Likelihood of W (t) is probability of seeing t1 , . . . , tn if W is true. Easier to work with log likelihood n Thinking about Smoothness What is an appropriate measure of smoothness for densities? x(t) = Ce W (t) dt Compare to Normal density f (t) = [LW (t)] dt 2 l (W |t1 , . . . , tn ) = i=1 W (ti ) log e W (t) Minimize the penalized negative log likelihood: n 1 2 2 e (t ) / 2 PENLOGLIK (W ) = i=1 W (ti )+n log e W (t) then W (t) = t 2 should be smooth LW (t) = D 3 W (t). Alternatively, LW (t) = D 2 W (t) exponential distribution is smooth useful for positive data. dt+ Usual comments about numerics apply. BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing Rough to Smooth Densities Every 10th beetle More Penalized Likelihood Sometimes squared error is not appropriate Binary data: yi {0, 1}; want to measure p(ti ) = P(yi = 1|t). Count data; want a Poisson intensity at each ti . Measurements of event times t1 , . . . , tn ; Poisson-process intensity can change over time. In each case we can write down a log likelihood of observed data given (some transformation of) W (t) and apply a smoothing penalty. These are discussed in the text, but not implemented in software. Mathematical theory justi es process. BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a> : Constrained Smoothing BTRY 6150: Applied Functional <a href="/keyword/data-analysis/" >data analysis</a&...

Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

Cornell - FDA - 2008
BTRY 6150: Applied Functional Data Analysis: From Data to Functions: Basis ExpansionsBTRY 6150: Applied Functional Data Analysis: From Data to Functions: Basis ExpansionsFrom Data To FunctionsHow do we go fromBasis ExpansionsFrom multiple lin
Cornell - FDA - 2008
BTRY 6150: Applied Functional Data Analysis: Models for DynamicsBTRY 6150: Applied Functional Data Analysis: Models for DynamicsReconsidering the Oil-Renery DataDierential EquationsSimple linear dierential equation model Dx(t) = x(t) + u(t) Co
Cornell - MULTIVARIA - 2008
BSCB 694: Theory of Multivariate StatisticsHomework 1Due: Thursday, Feb 281. Prove a Central Limit Theorem for S. Let xi (, ) be iid p-vectors with nite fourth moments. Then (n 1)1/2 vec(S ) N (0, ) in distribution for some . Demonstrate tha
Cornell - FDA - 2008
BSCB 6150: Applied Functional Data AnalysisHomework 2Due: Thursday, October 301. Monotone Smoothing and Infant Growth Data The onechild object in the fda library contains the length of a new-borns tibia, measured daily. We will be interested in
Cornell - FDA - 2008
BTRY 6150: Applied Functional Data AnalysisBTRY 6150: Applied Functional Data AnalysisApplied Functional Data AnalysisWhat is Functional Data?What are the most obvious features of these data?Venue: Tuesday/Thursday 11:40 - 12:55 WN 360 Lectu
Cornell - FDA - 2008
BTRY 6150: Applied Functional Data Analysis: FCCABTRY 6150: Applied Functional Data Analysis: FCCACanonical Correlations AnalysisMore Canonical CorrelationsAfter nding (a1 , b1 ) we can nd further canonical variates to maximize k (ak , bk ) =
Cornell - TNL - 7
Tristan Lefbure ePopulation Medicine &amp; Diagnostic Sciences College of Veterinary Medicine Cornell University Ithaca, NY 14850, USA Tel : (+1) 607 253-4228 Fax : (+1) 607 256-5608 TNL7@cornell.edu http:/www.people.cornell.edu/pages/tnl7Cursus unive
Cornell - TNL - 7
Tristan Lefbure ePopulation Medicine &amp; Diagnostic Sciences College of Veterinary Medicine Cornell University Ithaca, NY 14850, USA Phone: (607) 253-4228 Fax: (607) 256-5608 tnl7@cornell.edu http:/www.people.cornell.edu/pages/tnl7Education Ph.D. Ev
Cornell - ML - 593
September 2008 Min-Ah Lee CURRICULUM VITAEMin-Ah LeePostdoctoral Associate 220 Warren Hall Department of Development Sociology Cornell University Ithaca, NY. 14850 Phone: 607-342-6851 Email: ml593@cornell.eduEmployment2007 Present Postdoctoral
Cornell - CK - 275
Boston university Department of Economics Chulmin KimEC541 Monetary Theory &amp; Macroeconomics Professor Adrien Verdelhan Falll 2005Chulmin Kim (U38027154) 1) Preliminary tasks see the Excel File attached to the letter. The worksheets named Volcker
Cornell - JES - 259
Susan L. AverettDepartment of Economics and Business Lafayette CollegeAsia SikoraPolicy Analysis and Management Cornell UniversityIn Sickness and In Health: Marital Status and Adult HealthDependent Variable: Body Mass Index Independent Variabl
Cornell - JLL - 66
The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregisterThe current issue and full text archive of this journal is available at www.emeraldinsight.com/0960-4529.htmBeyond the friendly skies: an integr
Cornell - CS - 472
gm t dtx e d t h d m q h c f~ye f!f u}f|e cwvwmccT!3{fu7f tyes gain = no 130 yes 160 EQ no gain = gain = 190 positive positive negative positive negative good hacker publicationsm e d dmr x d m r r q t tr x d dt d x d t m e d dm
Cornell - CS - 472
t t d At t t tt t dt t t t t ttt l tjyqdUe ~ oj e td!Utjty d !h|4jt hsdh x7W t l do8td tj xU0|hj|tjWd !o8 n e q d { d { ~ q n eq { } q j u { q d 'yxts`shd|dts gsydqxsd !dts!tshWwx)wG!WsGtww q j {n {
Cornell - CS - 472
y d m d do } j m xm o } m !ozjaoiwvq|apo`w!kpdv8 c ozxu3r uq r lu cu!zj!aow4dapouvnrpiq kkrpa!a kd kevzm~po }'pwawdrweEm caou3pa4o x w j m l o d }l h j d h xwd h } j y d m d do } q xdj w j m j m } d q
Cornell - CS - 472
k e kq k k mj q s j k s e s kq rw q k k f~ se vuvkd!pf gfep|~!HvuA!vk|ei`d{ h! 0{vuC v}dhd}!|sv f j ~o s se s s f j ~o k w q k s k ~ e gk {di!Ym {|j{ie e|}|gk !{|es gk divu3!7 0{d 9dvs!w l4dA gikie Cgik4e}
Cornell - CS - 472
x y x v u g s r qph x9wwiet(aaigv u s r q o SEIm l wVq t98 pHRn9` 53 jd i h kw ii p e gw p 7R Ri(ah 0iX f d ed rr r r qr r ) ! ) 0&quot;d( y x v u g s r qph 59wwiet(aaigf ` cb IS ` 9edHE 4 a9&amp;Y ES QIGE 8 WXUVT7RPHFD CA@86 4 9B997
Cornell - CS - 472
v w v t s e q p ihf 9uugcr(YYge y E` a X E E`k X I E 'n7c9qHbb9fq7 E gS9caXRHT'T7pbbk t E t THgnBi'E 4 I d E r E S ` E S E y r` X E y S I E d u G Q r v dk dk X S I S a u k S v` gr&quot;g9 t P7h7Tc7b7b9TR97RjIg99b9X E u I S E k r d
Cornell - CS - 472
w x w v u g s r qph 9XXiet(aaigIf G S IE f b `S Q RBH9797HE y x w v u g s r qph 59XXiet(aaigf ` cb IS ` 9edHE 4 a9&amp;Y ES QIGE WXC UVT7RPHFD CA@86 4 9B99753 ) ! ) 2# ) % # ! &quot;0110('&amp;\$&quot;u x w v u g s r qph 9XXiet(aaigtI77V
Cornell - CS - 472
w x w u t f r q pig 9vvhds(`hf 97c79YHF'T9'T9Y 3 i e e S E D q E d Y X q E d q k I E I e E S q S E | e E b Y E } I E 0e7RPdbnRHT'Ec7m'9Vha j 9rRP d |S Q i e e Y Ea } e e HE`7HGo7c7TSBHcBi7h7QTS j A i7hRTH7e e I I S E | 9eHE9TSRrHTSc7q A P`dHs7e
Cornell - CS - 472
x y x v u g s r qph 9wwiet(aaig j~ }k { z x v u i srq o mk j n&quot;T|wywtt'np nlFi h g e d p h r fwfiiiaa7 ) !) 0&quot;&amp;w( y x v u g s r qph 59wwiet(aaigf ` cb IS ` 9edHE 4 a9&amp;Y W9WXUVT7RPHFD ES QIGE CA@86 4 9B99753 ) ! ) 2# ) % # !
Cornell - CS - 472
v w v t s e q p ihf 9uugcr(YYge y do ~ I uS E S 97b7YHE 'E e G e BX e 7 &quot;e 5ca e 'v}97bH7 I i S e I E G E g X ~ i` E g I d X |o Go IS X d g Ig E S E I do ` Xo G E | g g i S RB7z'zRn9| 9e YHpHRbF9z'HTh7TBX w e HpS e \$RI I iE ` uX HE d b`9Xz'7bon
Cornell - CS - 472
w x w u t f r q pig h9vvhds(`hft r q p o m SEIk j so 98 dnHRl9Y 53 i g f v hh i v ev i 7R\$ H&quot;(svh`g ( d qq q q pq q ) !) 0&quot;&amp;v( y x w u t f r q pig 59vvhds(`hfe Y ba IS Y 9dcHE 4 `9&amp;X ES QIGE W 9@ UVT7RPHFD CA@86 4 9B99753 ) !
Cornell - CS - 472
v w v t s e q p ihf 9uugcr(YYge p i h i p k f PHyPxR7eY(&quot;(57ug E` s a X Hb7c9x o d vq oq G E z } v o E | o I Q s I Q 9T7Brur'H7~7d A Hl{7lnBv z d X o Eq IS E s v S s w v I E o X I Q s dq o X o E m X p p k i h gh fh 9pHbly'xBT7(9tPutn7r
Cornell - CS - 472
y y w v h t s rqi 9xxpfu(bbph} {ey xf v u s q p d nml j hf e i~|z&quot;Twxtroo'ik igFd q i s pbb7 ) !) 0&quot;&amp;x( y w v h t s rqi 59xxpfu(bbphg a dc IS a 9feHE 4 b9&amp;` Y'WXUVT7RPHFD ES QIGE CA@86 4 9B99753 ) ! ) 2# ) % # !
Cornell - CS - 472
w x w u t d q p ige Avvfsr(hhfdS Q c E D I 77c S X FE ` X 7'E sy x w u t d q p ige 5Avvfsr(hhfdI ES QIGE c a AX b`'E 4 YS X&amp;W9C UVT7RPHFD C CB@86 4 AAA9753 ) ! ) 2# ) % # ! &quot;0110('&amp;\$&quot;t x w u t d q p ige Avvfsr(hhfdE ~G
ASU - APR - 2006
ARIZONA BOARD OF REGENTS ARIZONA STATE UNIVERSITY MEMORIAL UNION April 27 and 28, 2006 AGENDAThursday, April 2710:30 a.m. CALL TO ORDER, GREETINGS, AND ANNOUNCEMENTS FROM THE BOARD PRESIDENT CALL TO THE AUDIENCEThe Board shall not recognize any p
ASU - FEB - 2006
ARIZONA BOARD OF REGENTS ARIZONA STATE UNIVERSITY MEMORIAL UNION February 2 and 3, 2006 AGENDAThursday, February 21:00 p.m. 1:05 p.m. 1:20 p.m. CALL TO ORDER, GREETINGS, AND ANNOUNCEMENTS FROM THE BOARD PRESIDENT REPORT FROM THE PRESIDENT OF ARIZO
ASU - SEP - 2006
EXECUTIVE SUMMARYACTION ITEM:ABOR Capital Committee September 7, 2006 Agenda Item #3 Northern Arizona University Page 1 of 5Northern Arizona University requests Budget Increase for Conference Center and Parking Facility Northern Arizona Universi
ASU - AUG - 2006
ARIZONA BOARD OF REGENTS ARIZONA STATE UNIVERSITY MEMORIAL UNION August 10 and 11, 2006 AMENDED AGENDAThursday, August 101:30 p.m. CALL TO ORDER, GREETINGS, AND ANNOUNCEMENTS FROM THE BOARD PRESIDENT CALL TO THE AUDIENCE Per Board Policy 1-114, ti
ASU - APR - 2006
EXECUTIVE SUMMARY ACTION ITEM: ISSUE:ABOR Capital Committee March 28, 2006 Agenda Item #11 Northern Arizona University Page 1 of 4Request Project Approval, Wayfinding and Campus Entry ProjectNorthern Arizona University seeks Project Approval to
ASU - MAR - 2006
ARIZONA BOARD OF REGENTS UNIVERSITY OF ARIZONA STUDENT UNION MEMORIAL CENTER March 9 and 10, 2006 AGENDAThursday, March 91:00 p.m. CALL TO ORDER, GREETINGS, AND ANNOUNCEMENTS FROM THE BOARD PRESIDENT REPORT FROM THE PRESIDENT OF THE UNIVERSITY OF
ASU - AUG - 2006
ARIZONA BOARD OF REGENTS ARIZONA STATE UNIVERSITY MEMORIAL UNION August 10 and 11, 2006 AGENDAThursday, August 101:30 p.m. CALL TO ORDER, GREETINGS, AND ANNOUNCEMENTS FROM THE BOARD PRESIDENT CALL TO THE AUDIENCE ADOPTION OF ALL CONSENT ACTION ITE
ASU - JUN - 2006
ARIZONA BOARD OF REGENTS NORTHERN ARIZONA UNIVERSITY UNIVERSITY UNION June 22, 2006 AGENDAThursday, June 2210:30 a.m. CALL TO ORDER, GREETINGS, AND ANNOUNCEMENTS FROM THE BOARD PRESIDENT CALL TO THE AUDIENCE ADOPTION OF ALL CONSENT ACTION ITEMS AN
ASU - SEP - 2007
EXECUTIVE SUMMARYBoard of Regents Meeting September 27-28, 2007 Item #10 Page 1 of 5INFORMATION ITEM:PROPOSED AMENDMENT OF BOARD POLICY 4-311, ARIZONA STUDENT FINANCIAL AID TRUST TO CONFORM TO CHANGE IN ARIZONA LAW (FIRST READING)ISSUE:The
ASU - DEC - 2007
EXECUTIVE SUMMARY ACTION ITEM:Board of Regents Meeting December 6-7, 2007 Item # 16 Page 1 of 5PROPOSED AMENDMENT OF BOARD POLICY 4-311, ARIZONA STUDENT FINANCIAL AID TRUST TO CONFORM TO CHANGE IN ARIZONA LAW (SECOND READING)ISSUE:The Board i
ASU - SEP - 2006
Board of Regents Meeting September 27 &amp; 28, 2006 Item #32 Page 1 of 3EXECUTIVE SUMMARY ACTION ITEM: ASUs Biodesign Institute Project ISSUE: The Board is requested to review and approve ASUs Biodesign Institute project, for an anticipated expense of
ASU - MAR - 2007
Board of Regents Meeting March 8-9, 2007 Agenda Item #3 Page 1 of 5 EXECUTIVE SUMMARY ACTION ITEM: Approval of a Differential Tuition Fee for the Walter Cronkite School of Journalism and Mass Communication Program at Arizona State University for AY 2
ASU - EEE - 533
EEE 533: Semiconductor Device and Process Simulation Spring 2001 Lecture notes on mobility modeling Instructor: Dragica VasileskaDepartment of Electrical Engineering Arizona State UniversityEEE 533 - Semiconductor Device and Process Simulation1
ASU - CBALDWI - 1
Types of Urinary Management Systems, Part I There are many products available to manage urinary incontinence or urinary leakage. This article is Part I of a two-part article, which describes various products. The first article will discuss padding, e
ASU - JROBLES - 1
Learning Patterns to Answer Open Domain Questions on the WebDmitri RoussinovArizona State University P.O Box 873606 Tempe, AZ, 85287 (480)965-8488Jose RoblesArizona State University P.O Box 873606 Tempe, AZ, 85287 (480)965-3252dmitri.roussinov
ASU - AME - 2
Computer Vision and Image Understanding 95 (2004) 287316 www.elsevier.com/locate/cviuBayesian self-calibration of a moving cameraGang Qian and Rama Chellappa*Center for Automation Research, Institute for Advanced Computer Studies, University of M
ASU - AME - 08
Call for Papers Second workshop on Story Representation, Mechanism and Contexthttp:/ame2.asu.edu/SRMC08/in conjunction with acm mm 2008Vancouver, BC, Canada, October 31, 2008 http:/www.mcrlab.uottawa.ca/acmmm2008/ In this second SRMC workshop we
ASU - AME - 2
Call for Papers Second workshop on Story Representation, Mechanism and Contexthttp:/ame2.asu.edu/SRMC08/in conjunction with acm mm 2008Vancouver, BC, Canada, October 31, 2008 http:/www.mcrlab.uottawa.ca/acmmm2008/ In this second SRMC workshop we
ASU - AME - 2
Thanassis Rikakis Professor and Director Arts, Media and Engineering Program Arizona State University PO Box 873302 Tempe, AZ 85287-3302 (480) 965-0972 t.rikakis@asu.edu http:/ame2.asu.edu/faculty/thanassisAcademic Positions Professional Experienc
ASU - AME - 2
Multimedia and Hip-Hop for Experiential Education David Birchfield, Melanie West, Wilhelmina Savenye, Thanassis Rikakis1. Introduction In this article we present our recent work on two Student Centered Learning Environment projects that take experi
ASU - AME - 2
SMALLab: a mediated platform for educationDavid BirchfieldArts, Media and Engineering Arizona State University +1.480.965.3155Thomas CiufoArts, Media and Engineering Arizona State University +1.480.965.9438Gary MinyardTheatre for Youth Arizon
ASU - AME - 2
Sound and Interaction for K-12 Mediated EducationDavid Birchfield*, Thomas Ciufo*, Harvey Thornburg, Wilhelmina Savenye Arts, Media and Engineering, Arizona State University dbirchfield@asu.edu, thomas.ciufo@asu.edu Arts, Media and Engineering/Fult
ASU - AME - 2
Mediated Education in a Creative Arts Context: Research and Practice at Whittier Elementary SchoolAaron Cuthbertson, Sarah Hatton, Gary Minyard, Harper Piver, Christopher Todd, David Birchfield Arts, Media and Engineering Arizona State University PO
ASU - AME - 2
TRACKING THE PATH SHAPE QUALITIES OF HUMAN MOTION Kai Tu, Harvey Thornburg, Matthew Fulmer, and Andreas Spanias Arts, Media and Engineering Arizona State University 699 S. Mill Ave. Tempe, AZ 85287ABSTRACT We propose a probabilistic generative model
ASU - AME - 2
Learning Metaphor through Mixed-Reality Game Design and Game PlaySarah HattonArts, Media and Engineering Arizona State University Tempe, Arizona +1.480.965.3155David BirchfieldArts, Media and Engineering Arizona State University Tempe, Arizona +
ASU - AME - 2
Semantic Visual Templates: Linking Visual Features to SemanticsShih-Fu Chang William Chen Hari SundaramDept. of Electrical Engineering Columbia University New York New York 10027. E-mail: fsfchang,bchen,sundaramg@ctr.columbia.edu AbstractThe rapi
ASU - AME - 2
Bayesian Self-Calibration of a Moving CameraGang Qian and Rama ChellappaCenter for Automation Research and Department of Electrical and Computer Engineering University of Maryland College Park, MD 20742, U.S.A. {gqian, rama}@cfar.umd.eduAbstract.
ASU - AME - 2
w w { y3yy3 @ 4 &quot;&amp;d&amp; x{ { z H ! %x{ x1 9)8A\$)DB 9 9 5 ) 4 \$339&quot;@\$#H\$ B B x !9 x#&quot;! 6 d )#\$})8FE 5 B % 7 2 9 A 3@9g#\$8 z 3 8FE 5 )339&quot;@\$#H\$ z { % 7 !9 2 9 A B9 ( H 3 )@\$sB)H 4 l v } s ) drs3o )
ASU - AME - 2
94IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 14, NO. 1, JANUARY 2005Bayesian Algorithms for Simultaneous Structure From Motion Estimation of Multiple Independently Moving ObjectsGang Qian, Rama Chellappa, Fellow, IEEE, and Qinfen ZhengAbstract
ASU - AME - 2
Shreeharsh KelkarArts, Media and Engineering Program Department of Electrical Engineering, 699 S Mill Ave, Room 393DB, Arizona State University Tempe, Arizona 85287 RESEARCH INTERESTS Office: (480)-965-2548 Fax: (480)-965-0961 Cell: (623)-225-5704 s
ASU - AME - 2
Discovery of Blog Communities based on Mutual AwarenessYu-Ru Lin Hari Sundaram Yun Chi Jun Tatemura Belle TsengArts Media and Engineering Program Arizona State University NEC Laboratories America Cupertino, CA 95014e-mail: {yu-ru.lin, hari.sundar
ASU - AME - 2
1207, East 8th Street, #208 Tempe, AZ, 85281Bageshree Shevadeemail: Bageshree.Shevade@asu.edu Res: (480)-736-0423 Cell: (480)-277-3259 URL: http:/ame2.asu.edu/students/bageshreeSeeking a full-time position that will help develop my technical and
ASU - AME - 2
BLOG ANTENNA: SUMMARIZATION OF PERSONAL BLOG TEMPORAL DYNAMICS BASED ON SELF-SIMILARITY FACTORIZATIONYu-Ru Lin Hari Sundaram Arts Media Engineering, Arizona State University Tempe, AZ 85281Email: {yu-ru.lin, hari.sundaram}@asu.eduABSTRACTIn this
ASU - AME - 2
FacetNet: Community Discovery in Dynamic NetworksA Framework for Analyzing Communities and Their EvolutionsYu-Ru Lin1, Yun Chi2, Shenghuo Zhu2, Hari Sundaram1 and Belle Tseng31Arts, 2NECMedia and Engineering, Arizona State University Laboratorie
ASU - AME - 2
The Splog Detection Task and A Solution Based on Temporal and Link PropertiesYu-Ru Lin, Wen-Yen Chen, Xiaolin Shi, Richard Sia, Xiaodan Song, Yun Chi, Koji Hino, Hari Sundaram, Jun Tatemura and Belle Tseng NEC Laboratories America 10080 N. Wolfe Roa