| 3523 167th Ct NE #CC202, Redmond WA 98052, USA |
| dinoj@cs.uchicago.edu |
Tel: +1-773-7888287 |
| http://people.cs.uchicago.edu/~dinoj |
EDUCATION
| Dec 2007 |
Ph.D. Computer Science |
University of Chicago |
| Dec 2003 |
M.S. Computer Science |
University of Chicago |
| Dec 1999 |
M.Sc. Mathematics |
University of Zimbabwe |
| Dec 1997 |
B.Sc. Mathematics (Honours) |
University of Zimbabwe |
INTERESTS
| Applied Machine Learning | Scalable implementation of classification algorithms, Support Vector Machines,
Max-Margin Structured Learning for sequential data. |
| Computational Linguistics | Recognition of intonational patterns from acoustic features of speech, Determining the importance of phonemic contrasts and features in languages. |
| Information Visualization |
Visualization of astronomical data
Investigating different methods of visually representing the performance of machine learning algorithms for classification, dimensionality reduction, and clustering. Creation of informative glyphs. |
| Public Communication of Science | Creating animations and interactive visualization demos for researchers in different domains (machine learning, networking, astrophysics) to use in presentations.
Science Writing |
PROGRAMMING/COMPUTING KNOWLEDGE
|
C/C++, C#, Perl, Matlab, Scheme, Partiview, HTML, LaTeX, GNU/Linux.
|
PUBLICATIONS
Can Voice Quality help Mandarin Tone Recognition? by Dinoj Surendran and Gina-Anne Levow, Proceedings of ICASSP 2008.
"Dialog Act tagging with Support Vector Machines and Hidden Markov Models",
Dinoj Surendran and Gina-Anne Levow (Computer Science Department,
University of Chicago), Proceedings of Interspeech 2006.
"Classification of statement and question intonations in
Mandarin", Fang Liu (Linguistics Department, University of Chicago), Dinoj Surendran, and Yi Xu (Phonetics
Department, University College London). Proceedings of Speech
Prosody 2006, Dresden, Germany.
Interactive glyph-supplemented scatterplots for visualizing algorithm results. Dinoj Surendran. Companion Volume to the Proceedings of HLT-NAACL 2006.
"Quantifying the functional load of phonemic oppositions,
distinctive features, and suprasegmentals", Dinoj Surendran and Partha
Niyogi (Computer Science and Statistics Departments, University of Chicago), book chapter in
Nedergaard Thomsen, Ole (ed. fc.) Current trends in the theory of
linguistic change. In commemoration of Eugenio Coseriu (1921-2002).
Amsterdam & Philadelphia: Benjamins.
Estimation
of stellar atmospheric parameters from SDSS Ugriz Photometry.
Young Lee, Timothy Beers (Astronomy Department, Michigan State University), Coryn Bailer-Jones (Max-Planck-Institut fur Astronomie), Heidii
Newberg (Astronomy and Applied Physics Departments, Rensselaer Polytechnic Institute), Mark Subbarao (Astronomy Department, University of
Chicago), D. Surendran. Proceedings of the 207th Meeting of the
Astrononomical Association of America, Washington DC, January 2006.
Tone Recognition in Mandarin using
Focus.
Dinoj Surendran, Gina-Anne Levow, Yi Xu, Proceedings of the 9th European
Conference of Speech
Communication and Technology (Interspeech/ICSLP 2005)
Articulatory complexity, ambient frequency and functional load
as predictors of consonant development in children.. Stephanie
Stokes (Dept of Psychology, University of Reading) and Dinoj
Surendran. Journal of Speech and Hearing Research 48(3): 577-91, June 2005.
Visualizing
high dimensional datasets with Partiview. Dinoj Surendran and
Stuart Levy (NCSA). Poster Proceedings of the Tenth IEEE
Conference on Information Visualization, Austin, TX, 2004.
The functional load of tone
in Mandarin is as high as that of vowels. Dinoj Surendran and
Gina-Anne Levow. Proceedings of Speech Prosody
2004, Nara, Japan, pp. 99-102.
|
PRESENTATIONS
| May 2005 |
Automating
Visualization, "Visualization of Astrophysical Data: Bringing
Together Science, Art & Education" Workshop, Kavli Institute for Cosmological Physics, Chicago IL.
|
| Jan 2005 |
"3d Visualizations of current Astrophysical research for public explorations and classrooms". Demo with Randy Landsberg and Mark SubbaRao at "Gadgets and
Gizmos" Workshop, American Astronomical
Society Meeting, San Diego, CA.
|
| Dec 2004 |
"Partiview: a tool to visually analyze the perfomance of Dimensionality Reduction and Clustering algorithms". Demo
with Stuart Levy at Neural
Information Processing Systems
2004, Vancouver, BC.
|
| Oct 2004 |
Demo: "Stereo visualization of
astronomical data" with Randy Landsberg at the Great Lakes Planetarium Association
Conference, Detroit MI.
|
| Sep 2004 |
"Making pretty pictures for machine learning", Student session lecture Machine
Learning Summer School, Berder Island, France.
|
| Jun 2004 |
"Visualizing the Sloan Digital Sky Survey in 3d
Stereo" with Mark SubbaRao at the Physics
Education and Outreach Workshop, Aspen Center for Physics,
Aspen, CO.
|
| Mar 2002 |
"Language does not have to evolve
optimally" with Partha Niyogi at the Evolution of Language: Fourth International Conference, Harvard University, Cambridge, MA.
|
AWARDS
TEACHING EXPERIENCE
| Winter 2004 |
Lecturer, University of Chicago, Programming II (C++)
|
| Spring 2003 |
Lecturer, University of Chicago. Programming III (C++ and Data Structures)
|
| Summer 2002 |
Lecturer, University of Chicago. Programming I (Scheme).
|
| 2001-3 |
Teaching Assistant : Discrete
Mathematics, Artificial Intelligence, Programming. |
PROFESSIONAL EXPERIENCE
| Oct 2007 - |
Intern, Next Media Lab, Microsoft Research Redmond. Working on the
WorldWide Telescope project, processing terapixel astronomical images and data for global perusal.
|
| Winter 2007 |
Visualization Developer, Adler Planetarium and Astronomy Museum. Creating Python/Perl libraries for developing movie flight paths in Partiview and using them to create full-dome and 3d stereo movies of astronomical phenomena.
|
| Fall 2004 - Fall 2006 | Research
Assistant, Department of Computer Science, University of
Chicago. Working in Gina Levow's
NSF-funded Tone
and Intonation project. Applying information theoretic and machine
learning methods such as Support Vector Machines and
Max-Margin Structured Learning to large
data
sets of sequential prosodic information. Typically use C++, Matlab, Perl and Praat.
|
| Summer 2005 |
Summer Intern, Empirical Inference Department, Max
Planck Institute for Biological Cybernetics. Worked with Gunnar
Raetsch and Alexander Zien on Max-Margin
Generalized Hidden Markov Models. Sped up C++ implementation of existing Hidden Markov Model subroutine by over 20% in a gene-splicing algorithm by modifying Viterbi code to store less intermediate information.
|
| Spring 2005 |
Local Organizer, Chicago '05 Machine Learning
Summer School. Worked with John Langford, Yasemin Altun and others to arrange housing and practical tutorial sessions,
maintained website and Wiki.
|
| Summer 2002 |
Volunteer, Annual
Computational Neuroscience Meeting, Chicago, IL. Compiled the
Proceedings book.
|
| Summer 2001 |
Research Assistant, Mathematics and Computer Science Division, Argonne National Lab, Argonne, IL. Implemented a linear SVM in library-free C as part of a larger signal processing project.
|
| 1996-9 | Member
of the Board of Examiners, Zimbabwe Mathematics Olympiad. Worked with Erica Keogh (Statistics Department, University of Zimbabwe) to set questions for the nationwide competition for senior
high-school students. |
OUTREACH EXPERIENCE
| 2004-6 |
Volunteer, Cosmus Group. Created, documented, and freely distributed images, movies, and
interactive stereoscopic 3D animations for researchers and educators, mostly in the astronomical sciences.
|
| Dec 2005 |
Volunteer, Adler Planetarium. Helped create an
interactive planetarium dome show with Sky Skan Inc.
|
| Jun 2005 | Visualized time-varying data for the Markerless Motion Capture
project for Stefano Corazza (Mechanical Engineering Department,
Stanford University) using Matlab and Partiview.
|
| Sep 2004 | Created a visualization demo using Matlab and Partiview for Dimitri Achlioptas
(Microsoft Research) to use at his invited talk at the European
Conference for Machine Learning.
| | Aug 2004 | Created a visualization tool in Perl and
Partiview to help Matei Ripeanu (now faculty at University of British
Columbia) visualize experiments on the PlanetLab network. Images from this project are being used for an NSF Computing and Networking Sciences 2006 internal report.
|
| Summer 2004 |
Visualization Assistant, NSF-funded Kavli Institute for
Cosmological Physics. Worked with KICP Outreach Director Randy Landsberg astrophysicists Maximo Ave and Tokonatsu Yamamoto in the Pierre
Auger Observatory to create movies and interactive
animations of
cosmic
ray showers using AIRES (an existing simulation package), Perl, and Partiview. These have been used by researchers
and educators in conferences, classrooms, an upcoming documentary, and
in Astronomy magazine and the Pierre Auger website itself.
|
| Spring 2004 |
Visualized clustering patterns of handwritten digits
produced by the dimensionality reduction algorithm Laplacian Eigenmaps
of Mikhail Belkin (Ohio State) and Partha Niyogi.
|
| Mar 2004 |
Volunteer, Adler Planetarium and Astronomy
Museum. Aided Mark SubbaRao (University of Chicago and Adler) and Doug Roberts (Northwestern University and Adler) with some Scheme/GIMP programming to automate
the colorization of hundreds of stereo images from the NASA/JPL Mars Exploration Rovers to use in a stereoscopic 3D Mars exhibit at Adler.
|
| 2003-4 |
Intern, NSF-funded Museum
Presentation in Science program. Worked in a group of six graduatate students and two outreach professionals to create
stereoscopic 3D
astronomy-based exhibits for visitors to SciTech Hands-on Children's Museum
in Aurora, IL. Worked with Mark SubbaRao to create data for an
interactive model of the universe with data of hundreds of thousands of galaxies surveyed by the Sloan Digital Sky Survey created using Perl, Matlab, and Partiview. Installed this and other Stereo 3d exhibits at SciTech.
|
| 1997-9 |
Co-Founder and Vice Editor, Zimaths magazine, Mathematics Department,
University of Zimbabwe. Authored, edited, typesetted, and placed on web articles
for a magazine aimed at high school math students and teachers.
|
|
|