Kevin Duh

Signal, Speech, and Language Interpretation Lab
Department of Electrical Engineering
University of Washington
Seattle, WA 98195

Email: kevinduh (at) u.washington.edu
[My Resume (pdf)] [Publications] [Software] [Resources] [Recent Activities]

I have moved to NTT CS Labs.
Please visit my new homepage.

Graduate Research Assistant

Research Interests

I am a member of the Signal, Speech, and Language Interpretation (SSLI) Lab and my advisor is Katrin Kirchhoff. I am supported by the National Science Foundation (NSF) Fellowship (2005-2008). My current research focus is on the exciting and emerging area of: semi-supervised learning + ranking algorithms = semi-supervised ranking. The idea is to build better ranking functions (for information retrieval, speech recognition, or machine translation N-best list re-ranking, etc.) using a mix of labeled and unlabeled data.

Previously I have worked at several fun places, including Microsoft Research, NTT (Japan), and MIT Lincoln Labs.


Publications

  1. K. Duh and K. Kirchhoff, Learning to Rank with Partially-Labeled Data, ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2008)
  2. K. Duh and K. Kirchhoff, Beyond Log-linear Models: Boosted Minimum Error Rate Training for N-best Re-ranking, ACL 2008 (Short Paper)
  3. K. Duh, Ranking vs. Regression in Machine Translation Evaluation, Proc. of ACL 2008 Workshop on Statistical Machine Translation (Shared Evaluation Task), Columbus, Ohio, 2008.

  4. A. Axelrod, M. Yang, K. Duh, K. Kirchhoff, The UW Machine Translation System for ACL WMT 2008, Proc. of ACL 2008 Workshop on Statistical Machine Translation (Shared Translation Task), Columbus, Ohio, 2008

  5. K. Kirchhoff, J. Bilmes, and K. Duh, Factored Language Models - a Tutorial, UW-EE Technical Report, 2008-0004, Dept. of Electrical Eng., Univ. of Washington, Feb 2008.

  6. K. Kirchhoff, K. Duh, and C. Lim, The University of Washington Machine Translation System for IWSLT 2006, Proc. of the International Workshop on Spoken Language Translation, Kyoto, Japan, 2006. [slides]
  7. K. Duh and K. Kirchhoff, Lexicon Acquisition for Dialectal Arabic Using Transductive Learning, Proc. of EMNLP (Empirical Methods in Natural Language Processing), Sydney, Australia, 2006. [UW EE Tech Report]
  8. K. Kirchhoff, M. Yang, and K. Duh, Statistical Machine Translation of Parliamentary Proceedings Using Morpho-Syntactic Knowledge, Proc. of the TC-STAR Workshop on Speech to Speech Translation, Barcelona, Spain, 2006.
  9. S. Corston-Oliver, A. Aue, K. Duh, E. Ringger, Multilingual Dependency Parsing using Bayes Point Machines, Proc. of NAACL-HLT 2006, New York.
    (Slides of the talk I gave at Academia Sinica are available. The slides discusses parsing as structured classification and learning by Bayes Point Machines in more depth, but please refer to the paper for updated results.)
  10. K. Kirchhoff, D. Vergyri, J. Bilmes, K. Duh, and A. Stolcke, Morphology-based Language Modeling for Arabic Speech Recognition, Computer Speech and Language 20(4), 2006, pp.589-608
  11. K. Duh and K. Kirchhoff, Structured Multi-label Transductive Learning: a Case Study in Lexicon Acquisition, NIPS 2005, Advances in Structured Learning for Text and Speech Processing Workshop, Whistler, Canada, Dec 2005.
  12. K. Duh and K. Kirchhoff, POS Tagging of Dialectal Arabic: A Minimally Supervised Approach, 43rd Annual Meeting of the Assoc. for Computational Linguistics (ACL2005), Workshop on Computational Approaches to Semitic Languages, Ann Arbor, Michigan, USA, June 2005.
  13. K. Duh, Jointly Labeling Multiple Sequences: A Factorial HMM Approach, 43rd Annual Meeting of the Assoc. for Computational Linguistics (ACL 2005), Student Research Workshop, Ann Arbor, Michigan, USA, June 2005.
  14. C. Bartels, K. Duh, J. Bilmes, K. Kirchhoff, S. King, Genetic Triangulation of Graphical Models for Speech and Language Processing, Proceedings of Eurospeech/Interspeech 2005, Lisboa, Portugal, September 2005.
  15. D. Vergyri, K. Kirchhoff, K. Duh, A. Stolcke, Morphology-Based Language Modeling for Arabic Speech Recognition, International Conference on Spoken Language Processing (ICSLP/Interspeech 2004), October 2004.
  16. K. Duh and K. Kirchhoff, Automatic Learning of Language Model Structure, Proceedings of the 20th International Conference on Computational Linguistics (COLING-2004), Geneva, Switzerland, August 2004.
  17. K. Duh and K. Kirchhoff, Automatic Learning of Language Model Structure (Extended version of COLING-2004 paper), UW-EE Technical Report 2004-0014, Dept. of Electrical Engineering, University of Washington, April 2004.
  18. K. Duh, Least Squares Estimation of GPS L1-L2 Satellite and Receiver Bias, MIT Lincoln Laboratory Project Report, Nov 2003.

Downloadable Software


Resources

Some notes, tutorials, and presentations (PDF or HTML format) that may be of interest:

Previous Projects


Recent Activities

Links

View Kevin Duh's profile on
LinkedIn
"A language is a dialect with an army and a navy" -- attributed to Max Weinreich, linguist (1883-1969)
"Your idea is crazy, but not crazy enough to be true" -- Niels Bohr, physicist (1885 - 1962)
"What does not destroy me makes me stronger" -- Friedrich Nietzsche, philosopher (1844-1900)
"The mind that faces and defeats one opponent is the same mind that faces and defeats a thousand." -- Miyamoto Musashi, swordsman (1584-1645)
"The future is here. It is just not evenly distributed" -- William Gibson, author (1948- )

Back to the SSLI Lab homepage
Copyright 1995-2008 University of Washington, College of Engineering - SSLI Lab
(Most Recent Update: Apr 15, 2008)