* Harr Chen Computer Science and Artificial Intelligence Laboratory Massachusetts Institute of Technology The Stata Center, 32-G494 32 Vassar Street Cambridge, MA 02139 harr at mit dot edu / harrchenone (AIM) / harrchen at msn dot com (MSN) / harr at gmail dot com (GTalk)

I am a graduate student advised by Professor David Karger and Professor Regina Barzilay. Perhaps you would like to look at my résumé?

Research

• Harr Chen, S.R.K. Branavan, Regina Barzilay, David R. Karger. PREPRINT: Global Models of Document Structure Using Latent Permutations. To appear in Journal of Artificial Intelligence Research.
• S.R.K. Branavan, Harr Chen, Luke Zettlemoyer, Regina Barzilay. Reinforcement Learning for Mapping Instructions to Actions. To appear in Proceedings of ACL/IJCNLP 2009.
• Harr Chen, S.R.K. Branavan, Regina Barzilay, David R. Karger. Global Models of Document Structure Using Latent Permutations. In Proceedings of NAACL/HLT 2009.
• S.R.K. Branavan, Harr Chen, Jacob Eisenstein, Regina Barzilay. Learning Document-Level Semantic Properties from Free-text Annotations. Journal of Artificial Intelligence Research Volume 34 (2009).
• Kuang Chen, Harr Chen, Neil Conway, Heather Dolan, Joseph Hellerstein, Tapan Parikh. Improving Data Quality with Dynamic Forms. Demo abstract in Proceedings of IEEE/ACM ICTD 2009.
• S.R.K. Branavan, Harr Chen, Jacob Eisenstein, Regina Barzilay. Learning Document-Level Semantic Properties from Free-text Annotations. In Proceedings of ACL/HLT 2008.
• Harr Chen, David R. Karger. Less is More: Probabilistic Models for Retrieving Fewer Relevant Documents. In Proceedings of ACM SIGIR 2006. Slides: [ppt]
• Raman Chandrasekar, Harr Chen, Simon Corston-Oliver, Eric Brill. Subwebs for Specialized Search. In Proceedings of ACM SIGIR 2004.

• Harr Chen, Adwait Ratnaparkhi, Sonja Knoll, Hsiao-Wuen Hon. System and method for using anchor text as training data for classifier-based search systems. US Patent No. 7480667. And yes, the USPTO managed to reverse my name in the patent issue, even though it was correct in the published application. :(
• Harr Chen, Raman Chandrasekar, Simon Corston-Oliver, Eric Brill. Building and using subwebs for focused search. US Patent No. 7392278.
• Simon Corston-Oliver, Raman Chandrasekar, Harr Chen. Machine-learned approach to determining document relevance for search over large electronic collections of documents. US Patent No. 7287012.

Personal

• interested primarily in information retrieval and natural language processing
• affiliated with the Haystack project
• funded by the Office of Naval Research National Science Foundation
• aka harrbar, harrchenemy
• a sage who sits above Asgard, the other two being Jafnharr and Thridi
• the pirate's favorite computer scientist (arrrgh harrrrr!)
• way too fond of HTML lists
• defined on Urban Dictionary (though I'm not actually ABC)
• redirected on Wikipedia to the Norse God Odin

• a student of the Sloan School of Management Financial Technology Option
• a Credit Suisse proprietary trading (New York) and equity derivatives trading (Hong Kong) summer associate
• a Google research intern
• a Microsoft full time software engineer
• a University of Washington undergraduate
• a graduate of the UW Early Entrance Program
• a student at Margaret McKenny elementary school, Olympia, WA (until last year my third grade project on penguins, complete with my recorded audio commentary, was still online - I was such a cute little kid)
• born on April Fool's Day in 1987 in the beautiful port city of Qingdao, Shandong Province, P.R. China; many westerners will have heard of our city's most famous export, Tsingtao Beer

• an Erdos Number of four, putting me in distinguished company

• Brazil
• Canada
• P.R. China (country of birth)
• France
• Hong Kong
• Italy
• Germany
• Macau
• The Netherlands
• The Philippines
• United Kingdom
• United States (country of residence)