Damian Eads

Damian Eads
Ph.D. Student, Computer Science
Jack Baskin School of Engineering
University of California at Santa Cruz

Menu:

• Home
• CV/Resume
• Papers
• Coursework
• Education
• Software
• Google Scholar
• Data

• Facebook
• LinkedIn

[Printable Version]

Damian Eads

Jack Baskin School of Engineering University of California, Santa Cruz 1156 High Street Santa Cruz, California 95064 USA

Space and Remote Sensing Sciences (ISR-2) Los Alamos National Laboratory MS B244 Los Alamos, NM 87545

Note: This is my unabridged resume. A one page, specialized resume tailored for a specific venue is available by request.

Objective

• To better understand the physical world through graduate study and research.
• To serve society by developing algorithms to tackle real world problems tested with rigorous statistical analyses or proof-based methods.
• To strive to share my implementations free of charge without restriction, and to promote software freedom in research institutions.

Research Interests

• machine learning, statistics, computer vision, analysis of remote sensing data, engineering scientific tools, information retrieval, graph theory, optimization theory, foreign affairs, government studies

Education

• 2005-: Doctor of Philosophy (Ph.D.), Computer Science
  University of California, Santa Cruz
  Adviser: Dr. David Helmbold
  
  Expected Date of Advancement to Candidacy: December of 2008
  Expected Date of Degree: June of 2010
  
• 1999-2004: Bachelor of Science (B.Sc.), High Honors, Computer Science
  Golisano College, Rochester Institute of Technology
  
  Minors: Mathematics and Psychology
  GPA: 3.62/4.00, Major GPA: 3.74/4.00
  
• 1995-1999: High School Diploma, Ridgefield High School, Ridgefield, CT

Skills

• Computer Languages: Java, C, C++, SQL, Clipper, PROLOG, Common Lisp, Scheme, Ocaml, NJML, Python, R, MATLAB, LaTeX, bash, Visual Basic, PHP4+5, HTML, ColdFusion, VAX Assembly, x86 (pre-386), Ruby*, Maple*, Maxima*, J*
• Operating Systems: Solaris, Linux, Windows
• Software: Rational Rose, StarOffice, MS Office, Exceed, RCS, CVS, make, automake, autoconf, lex, bison.

[*] currently learning

Relevant Graduate Coursework (13 classes total)

• Advanced Operating Systems, Programming Semantics, Research and Teaching, Storage Systems, Analysis of Algorithms, Advanced Computer Security, Information Retrieval, Optimization Theory, Machine Learning, Bayesian Statistics, Combinatorial Algorithms

Relevant Undergraduate Coursework (55 classes total)

• Math: Calculus I-IV, Discrete Mathematics I & II, Probability and Statistics, Matrix Algebra, Graph Theory, Combinatorics
• Science: Chemical Principles I & II, Organic Chemistry Principles, Environmental Applications of Remote Sensing, Astronomical Imaging
• Computer Science: Computer Science I-IV, Software Engineering, Artificial Intelligence, Computer Vision, Programming Language Concepts, Introduction to Digital Design, JRMS, Language Processors, Genetic Algorithms, Operating Systems I, Computer Organization, Computer Science Theory, Data Communication and Networking, Abstractions and Practicalities, Virtual Theatre
• Relevant Liberal Arts: American History, Writing and Literature I & II, Abnormal Psychology, Psychology of Personality, Psychology Research Methods, Sociology

Employment

• Graduate Student Researcher, Department of Computer Science, Jack Baskin School of Engineering, University of California at Santa Cruz
  January 2006-June 2006
  September 2006-June 2007
  July 2007 (no activity)
  August 2007-January 2008 (leave of absence)
  January 2008-March 2008 (independent study, remotely from LANL)
  April 2008-June 2008
  • Developing ``Purple'': a new algorithm for rapidly learning robust object detectors in unstructured example spaces. (May 2008-)
  • Studying theoretical aspects of sequences exemplars. Proved a set of useful lemmas, laying the ground for future work. (April 2008-)
  • Building a new machine learning tool suite for Python. (February 2008-)
  • Developing algorithms for classifying, clustering, and retrieving time series transients in large data sets. (2006-present)
  • Developed prototype algorithms for the adaptive migration of files in large, non-hierarchical storage systems using machine learning. (2006)
  • Investigated the use of time series forecasting to predict file system workloads. (2006)
• Staff Research Assistant, Machine Learning Team, Space and Remote Sensing Sciences Group, Los Alamos National Laboratory June 2005-September 2005 (full time)
  September 2005-November 2005 (on leave, Ph.D. studies)
  December 2005 (visit)
  January 2006 (on leave, Ph.D. studies)
  February 2006 (visit)
  February 2006-March 2006 (on leave, Ph.D. studies)
  March 2006 (visit)
  April 2006-June 2006 (on leave, Ph.D. studies)
  June 2006-September 2006 (full time)
  September 2006-December 2006 (on leave, Ph.D. studies)
  December 2006 (visit)
  January 2007-June 2007 (on leave, Ph.D. studies)
  August 2007-April 2008 (full-time)
  April 2008-June 2008 (on leave, Ph.D. studies)
  June 2008-October 2008 (full-time)
  • Made substantial and critical contributions to an important project in 2008. Met many sudden deadlines with quality and innovative solutions to problems that were deemed unlikely to be solved by the time needed.
  • Participated in a competition sponsored by a US government agency. (2008)
  • Visited US government sponsors and collaborators in DC and other areas. (October 2007-April 2008)
  • Implemented Bacon: an algorithm for rapid object detection in panchromatic imagery. Demonstrated performance of algorithm on several large data sets. Developed production release that was ported to both Windows and Linux. Responsible for ground-up design and implementation of large scale system. (October 2007-March 2007)
  • Implemented Pigs: an algorithm for extracting features from images and signals using grammars. (October 2007)
  • Developed algorithms for clustering variable-length and irregularly sampled time series. (December 2007)
  • Investigated new algorithms for generating representative time-series prototypes. (December 2007)
  • Wrote a reference manual on the LANL Sparse Image Format (SIF). (October 2007)
  • Instrumental in the development of Genie Pro, a commercial algorithm for pixel classification used for many geospatial and remote sensing applications. Genie Pro placed first against three other competitors in an evaluation sponsored by a US government agency.
  • Investigated the use of different feature selection algorithms for Zeus, a system for generating time series classifiers.
  • Received the Distinguished Copyright of the Year Award from LANL in both 2005 and 2006 for my extensive work on developing technologies that were successfully commercialized into shrink-wrapped products.
  • Studied non-linear time series analysis and simple algorithms for detecting chaotic time series from surrogates.
  • Applied existing algorithms to challenging problems of interest to government sponsors.
  • Participated in a reading group on convex optimization. (August 2005)
  • Developed ideas for future dissertation work. (June-August 2006, September 2007-March 2008)
  • Member of LANL's LGBTI Diversity Task Force.
• Postbaccalaureate Technician, Machine Learning Team, Space and Remote Sensing Sciences Group, Los Alamos National Laboratory June 2004-June 2005
  • Authored third version of Zeus, a system for generating time series classifiers. Designed an algorithm for grammar-guided feature extraction that has been used by colleagues for numerous tasks including learning edge detectors and pixel classifiers.
  • Integrated grammar-guided feature extraction algorithm into Genie Pro, a commercial version of GENIE.
  • Designed a file format (SIF) for storing sparse images. Implemented an I/O library to manipulate SIF files.
  • Gave briefings and demonstrations of algorithms I developed to sponsors.
  • Participated in a Pattern Classification reading group.
• Undergraduate Student, Machine Learning Team, Space and Remote Sensing Sciences Group, Nonproliferation and International Security Division, Los Alamos National Laboratory, U.S. Department of Energy
  June 2001-November 2001
  February 2002-March 2002
  July 2002-May 2003
  • Created a machine learning tool called Zeus to solve a generic set of pattern recognition problems involving pixel classification, time series classification, time series forecasting, and image classification. This system has been used to tackle difficult real-world problems involving remote sensing and geophysical modeling.
  • Developed source for the GENIE (Genetic Imagery Exploitation) Pixel Classification Project. GENIE uses evolutionary computation to evolve imagery algorithms for rapid feature extraction in hyperspectral, multispectral, and panchromatic imagery. The GENIE team was a recipient of the R & D 100 and LANL Distinguished Performance awards. The tool set has also been positively mentioned in testimonies given to the U.S. Senate Armed Services Committee and U.S. Senate Energy and Natural Resources Committee.
  • Developed ChroMX, an exploitation tool used by the GENIE team to compile IDL chromosomes into ANSI-standard C programs.
  • Created Magic Carpet, a software application that manages learning sessions from three different machine learning systems within Aladdin-- a software system used to produce training data for pixel classifiers.
  • Collaborated with hardware engineers to develop software for generating training data for ingestion in a video object detection system. Also created a visualization tool to enable the user to see the progress of the system as it learned.
• Student Assistant, Department of Computer Science, Rochester Institute of Technology
  December 2000-February 2001
  December 2003-February 2004
  • Provided assistance to both undergraduate and graduate students enrolled in Professor Hans-Peter Bischof's JRMS course.
  • Developed course material for Professor Strout's Programming Language Concepts (4003-450, 4003-708) class. Assisted in the grading of labs and projects for three undergraduate and graduate sections (~120 students).
  • Delivered 2-hour substitute lecture on advanced LISP programming.
• Writer/Programmer, Department of Computer Science, Rochester Institute of Technology
  September 2000-January 2001
  • Contributed to the development of a tutorial for the Java Reliable Multicast Library (JRMS).
• Lab Instructor, Department of Computer Science, Rochester Institute of Technology
  September 2000-March 2004
  • Responsible for instruction of a weekly lab section for Computer Science I, two sections of Computer Science II, and five sections of Computer Science IV.
  • Responsible for the grading of weekly lab assignments.
  • Provided weekly debriefings for students to identify areas of difficulty and suggest ways to improve.
  • Held office hours each week to offer students extra help and support.
  • Collaborated with instructors and faculty to suggest improvements in the content of the CS I, II, and IV courses.
• Programmer, One Web Source, Ridgefield, CT
  May-August 2000
  • Developed an e-commerce application in ColdFusion for a promotional products company.

Peer Review

• Reviewer, IEEE Transactions on Evolutionary Computation (2006)
• Reviewer, Journal of Machine Learning Research (JMLR) (2008)

Conference and Journal Papers (13)

• 13. D. Eads and E. Rosten. "Experiences Using SciPy for Computer Vision Research." Proceedings of the 2008 SciPy Conference. California Institute of Technology. August 20, 2008. (6 pages). LA-UR 08-05862. Pending print.
• 12. J. S. Bloom, D. L. Starr, N. R. Butler, P. Nugent, M. Rischard, D. Eads, and D. Poznanski. "Towards a Real-time Transient Classification Engine." Astronomische Nachrichten. Volume 329. Issue 3. March, 2008. pp 284-7. In print.
• 11. C. Maltzahn, N. Bobb, M. W. Storer, D. Eads, S. A. Brandt, and E. L. Miller, "Graffiti: A Framework for Testing Collaborative Distributed Metadata," Journal of Informatics, 21. March, 2007. pp. 97-111. In print.
• 10. P. Zigoris and D. Eads. "Unsupervised Learning of Tree Alignment Models for Information Extraction." The International Conference on Data Mining (ICDM) Workshop on Ontology Mining and Knowledge Discovery from Semi-Structured Documents. Hong Kong, China. December, 2006. pp. 45-49. In print.
• 9. D. Eads, K. Glocer, S. Perkins, J. Theiler. "Grammar-guided Feature Extraction for Time Series Classification." Los Alamos National Laboratory Technical Report LA-UR-05-4487. June, 2005.
• 8. N. Bobb, D. Eads, M. W. Storer, S. A. Brandt, C. Maltzahn, E. L. Miller. "Graffiti: A Framework for Testing Collaborative Metadata." Proceedings of the 2006 Workshop on Distributed Data and Structures (WDAS 2006). Santa Clara, CA. In print.
• 7. K. Glocer, D. Eads, and J. Theiler. "Online Feature Selection for Pixel Classification." Twenty-second International Conference on Machine Learning (ICML-05). Bonn, Germany. pp. 249-256. In print.
• 6. S. Perkins, K. Edlund, D. Esch-Mosher, D. Eads, N. Harvey, and S. Brumby. "Genie Pro: Robust Image Classification using Shape, Texture, and Spectral Information." SPIE Conference on Defense and Security. Orlando, FL. April, 2005. pp. 139-148. In print.
• 5. R. Porter, D. Eads, D. Hush and J. Theiler. "Weighted Order Statistic Classifiers with Large Rank-Order Margin." Twentieth International Conference on Machine Learning (ICML-03). Washington, DC: August, 2003. pp. 600-607. In print.
• 4. D. Eads, S. Williams, J. Theiler, R. Porter, N. Harvey, S. Perkins, S. Brumby, and N. David. "A Multimodal Approach to Feature Extraction for Signal and Pixel Learning Problems Using Evolutionary Computation." Sixth Conference on the Applications and Science of Neural Networks, Fuzzy Systems, and Evolutionary Computation. Symposium on Optical Science and Technology of the 2003 SPIE Annual Meeting. San Diego, CA: August, 2003. SPIE Proc. 5200. pp. 79-90. In print. (LA-UR-03-0086)
• 3. D. Eads, S. Davis, S. Perkins, J. Theiler, J. Ma, R. Porter. "Genetic Algorithms and Support Vector Machines for Time Series Classification." Fifth Conference on the Applications and Science of Neural Networks, Fuzzy Systems, and Evolutionary Computation. Symposium on Optical Science and Technology of the SPIE Annual Meeting. Published in SPIE Proc. 4787. Seattle, WA: July, 2002. pp. 74-85. In print. (LA-UR-01-6729) and (LA-UR-02-6212)
• 2. J. J. Szymanski, S. P. Brumby, P. Pope, D. Eads, D. Esch-Mosher, M. Galassi, N. R. Harvey, H. D. W. McCulloch, S. J. Perkins, R. Porter, J. Theiler, A. C. Young, J. J. Bloch, and N. David. "Feature Extraction from Multiple Data Sources using Genetic Programming." 2002 SPIE AeroSense Conference. Algorithms and Technologies for Multispectral, Hyperspectral, and Ultraspectral Imagery VIII. Orlando, FL: April, 2002. Published in SPIE Proc. 4725. pp. 338-445. In print. (LA-UR-02-1551)
• 1. J. Binder, H. P. Bischof, J. Coles, D. Eads, C. Fagan, and J. Myers. "Java Reliable Multicast Service (JRMS): A Tutorial." January, 2000.

Class Papers (11)

• 11. D. Eads. "Learning Object Detectors." June, 2008. CMPS 297.
• 10. D. Eads. "Optimal Sequence Exemplars." June, 2008. CMPS 297.
• 9. D. Eads. "A Report on the Chestnut Scientific Library." June, 2008. CMPS 297.
• 8. D. Eads. "Object Detection in Panchromatic Imagery" March, 2008. CMPS 297.
• 7. D. Eads. "CASCC: A New Algorithm for Time Series Classification." December, 2006. CMPS 211.
• 6. D. Eads. "An Introduction to Nonlinear Programming." November, 2006. ISM 206.
• 5. D. Eads and P. Zigoris. "Machine Learning for Information Extraction." June, 2006. ISM 260.
• 4. D. Eads. "The Trust System for the Simply-typed Lambda Calculus." June, 2006. CMPS 223.
• 3. D. Eads. "Using Machine Learning for Adaptive Placement of Files in Non-hierarchical Storage Systems." March, 2006. CMPS 229.
• 2. J. Johnson and D. Eads. "An Implementation of a Language with Multiple Dispatch." December, 2005. CMPS 203.
• 1. D. Eads. "MRAM: Exploring Alternative Data Structures for Low-level Filesystem Management." December, 2005. CMPS 221.

Memberships

• American Academy of Arts and Sciences (AAAS)
• Association for Computing Machinery (ACM)
• Institute for Electrical and Electronics Engineers, Inc. (IEEE)
• Society of Photo-Optical Instrumentation Engineers (SPIE)

References

Specialized references are available upon request.

Graduate School

• David Helmbold, Ph.D., Professor, U.C. Santa Cruz
• Ethan Miller, Ph.D., Associate Professor, U.C. Santa Cruz
• Joshua Bloom, Ph.D., Assistant Professor, U.C. Berkeley
• Jarrod Millman, Director of Computing, H.W. Neuroscience Inst., U.C. Berkeley

Employment

• James Theiler, Ph.D., Laboratory Fellow, Los Alamos National Laboratory
• Ed Rosten, Ph.D., Postdoc, Los Alamos National Laboratory
• Simon Perkins, Ph.D., Vice President for Research, Geospace, Inc.
• Reid Porter, Ph.D., Staff Member, Los Alamos National Laboratory
• Mark Galassi, Ph.D., Staff Member, Los Alamos National Laboratory
• Nancy David, Ph.D., Team Leader, Los Alamos National Laboratory
• Neal Harvey, Ph.D., Staff Member, Los Alamos National Laboratory
• Kim Edlund, M.Sc., Staff Member, Los Alamos National Laboratory
• Andy Fraser, Ph.D., Staff Member, Los Alamos National Laboratory
• Steven Brumby, Ph.D., Staff Member, Los Alamos National Laboratory
• Abram Jacobson, Ph.D., Professor, Earth and Space, University of Washington
• Douglas Alde, Ph.D., Staff Member, Los Alamos National Laboratory
• Jeffrey Bloch, Ph.D., Staff Member, Los Alamos National Laboratory
• Laura Boucheron, Ph.D., Visiting Scientist, Los Alamos National Laboratory
• Dan Starr, B.Sc., Software Engineer, University of California, Berkeley
• Doreen Fleming, Owner, One Web Source

Undergraduate

• Sean Strout, M.Sc., Lecturer, Computer Science, RIT
• Roger Gaborski, Ph.D., Associate Dean, College of Computing, RIT
• Joe Geigel, Sc.D., Associate Professor, Computer Science, RIT
• Paul Wilson, Ph.D., Professor, Mathematics, RIT
• Peter Anderson, Ph.D., Professor Emeritus, Computer Science, RIT
• Jonathan Coles, M.Sc., Ph.D. Student, Theoretische Physik Institut, Universität Zürich

Software Projects (34 major projects)

• aladdin: helped maintain LANL's Aladdin GUI, a tool for creating training data for pixel classification algorithms. (2001, retired)
  Los Alamos Computer Code #: LA-CC-01-XXX
• bacon: an algorithm for rapid detection of objects in panchromatic imagery. (October, 2007, founder)
  Los Alamos Computer Code #: LA-CC-08-090
• chestnut: a suite of machine learning and data mining algorithms written in Python. (April, 2008, founder)
• dzgs: a prototype grammar-guide feature extraction system written in MATLAB and Java. Used for learning signal classifiers and edge detectors. (2004, founder) Los Alamos Computer Code #: LA-CC-05-066
• Fast Time Series (FTS): efficient implementations of Dynamic Time Warping (DTW) algorithms for Numpy/Scipy. (2007, founder)
• Genie Pro: A machine learning tool for learning pixel classifiers for imagery data sets. (2004, major co-contributor)
  Los Alamos Computer Code #: LA-CC-04-XXX, LA-CC-07-014
  Licensed to:
  • Observera, Inc.
  • Aperio Technologies, Inc.
• hcluster: A hierarchical clustering and dendrogram plotting module written in Python and C. (September, 2007, founder)
• itemizer: A tool for learning the structure and layout of HTML pages for automated information extraction. (2006, co-founder)
• libcvd: Developing a Python interface to the Cambridge Video Dynamics library. (2008, junior co-contributor)
• magic carpet: A tool for managing batch machine learning jobs. (2001, retired, founder)
• matdoc: A program for generating HTML, LaTeX, and matlab-help documentation files from MATLAB and Octave programs. (2003, founder)
  Los Alamos Computer Code #: LA-CC-03-049
• mfitsio: A FITS (Flexible Image Transport System) reading library for the MATLAB environment. (2002, founder)
  Los Alamos Computer Code #: LA-CC-02-086
• mji: An interface for invoking MATLAB commands from a Java program. (2003, founder)
  Los Alamos Computer Code #: LA-CC-03-050
• mmls: A small python library containing sundry machine learning algorithms I wrote during my graduate work. (2007, founder)
• MultiCruz: A functional programming language for multiple method dispatch. (2005, founder)
• osu-svm: A support vector machine toolbox for the MATLAB environment. Revised API documentation. (2003, contributed to documentation)
• pigs: A Python/Scipy software system for optimizing graph-structured feature extraction algorithms using grammars in the Python/Scipy environment. (October, 2007, founder) (Published in LA-UR 08-2384)
  Los Alamos Computer Code #: LA-CC-08-090
• polenta: A functional and object oriented programming language written in Java. (2001, founder)
• purple: A new object detection system using unstructured boosting. (2008, founder)
• py-victory: A rewrite of the victory package in Python/Django. (2007, founder)
• rbtree-lifs: Adapted UCSC's LiFS file system to use Red-Black Trees to manage allocation and file offset indexing. (2005, founder of sub-project)
• ritfob: a Java library for interfacing with Ascension's Flock of Birds device. (2004, founder)
• ritglove: a Java library for interfacing with the 3DT glove. (2004, founder)
• ritopt: An option parsing library for the Java programming language. (2001, founder)
• scipy: A tool suite for scientific computation in Python. (junior developer -- one of dozens of co-contributors)
• sif: A file format for storing sparse image data. (2004, founder)
  Los Alamos Computer Code #: LA-CC-06-066
  Licensed under the terms of the LGPL
• s-trust: A basic implementation of the trust calculus written in Scheme. (2006, founder)
• tsalign: An extension to Scipy for performing classification of time series using DTW and other algorithms. (2006, founder)
• tsp-viz: A Traveling Salesperson Problem (TSP) visualization program written in Java. (2002, founder)
• v-aladdin: An extension to LANL's Aladdin GUI for handling video streams. (founder of sub-project, 2003)
• victory: PHP software for managing door-to-door political canvassing campaigns. (2004 election season, co-founder)
• ZGS (Zeus Grammar System): A C++ library for randomly generating computer programs from grammatical descriptions. (2005, founder)
  Los Alamos Computer Code #: LA-CC-05-XXX.
  Licensed to Observera Inc. and Aperio Technologies
• Zeus: A time-series classification system written in MATLAB. (2005, founder) Los Alamos Computer Code #: LA-CC-05-066.

Teaching Experience (15 sections)

• Taught 4003-231 Computer Science I Laboratory (One Section), RIT
• Taught 4003-232 Computer Science II Laboratory (Two Sections), RIT
• Taught 4003-334 Computer Science IV Laboratory (Five Sections), RIT
• Graded 4003-450 Programming Language Concepts (Three Sections), RIT
• Graded 4005-709 Programming Language Concepts (Three Sections), RIT
• Delivered one substitute lecture for 4003-450 and 4005-709, RIT
• TA'ed 4003-590 Introduction to JRMS (Part of One Section), RIT
• TA'ed 4005-779 JRMS (Part of One Section), RIT

Community Service

• Vice President, Ridgefield High School (RHS) Tiger TV, 1998-1999
• Secretary, Amnesty International (RHS Chapter), 1997-1999
• Member, LANL LGBTI Diversity Task Force, 2004-2008
• Santa Fe Pride on the Plaza, 2005, 2006
• Organizer, 1st LANL LGBTI Student/Postdoc Picnic, 2005
• Donations to various charities (Doctors without Borders, Amnesty International, NM State Police, National Children's Leukemia, Human Rights Campaign, San Francisco AIDS Marathon)
• Sue Chef (a few occasions), Homeless Shelter, Santa Fe, MN

Awards and Honors

• Marquis Who's Who in America, 2009, 63rd Edition
• Golden Key International Honour Society
• Dean's List (for Eight Quarters), RIT
• Member of team awarded R&D Magazine's R&D 100 Award (2002)
• Member of team awarded LANL's Distinguished Performance Award (06/2002)
• LANL Nonproliferation and International Security Scholarship (09/2003)
• LANL SPOT Award (08/2004)
• LANL Acheivement Award (LAAP) (09/2005)
• LANL 2005 Distinguished Copyright of the Year Award (03/2006), Genie Pro
• LANL Five Years of Service Award (06/2006)
• LANL Acheivement Award (LAAP) (09/2006)
• LANL 2006 Distinguished Copyright of the Year Award (04/2007), Genie Pro
• LANL Copyright Honorable Mention (05/2008), Sparse Image Format (SIF)
• LANL SPOT Award (06/2008)

Citizenship

• United States (by birth)

Copyright © Damian Eads 1999-2008. All Rights Reserved. Design by Andreas Viklund.