About Me

IMG_20131123_212633 (1)I’m a Senior Research Scientist at Semantic Machines, working to build dialogue systems that you can actually have a conversation with.

Previously, I received my PhD in EECS at Berkeley, working in the Berkeley NLP Group with Dan Klein.  Before that, I was an undergrad in Symbolic Systems at Stanford University, working in the Stanford NLP Group with Dan Jurafsky and Chris Manning.

Email: dlwh at cs.[[berkeley]].edu

Publications

  • [PDF] W. Chang, C. Cathcart, D. Hall, and A. Garrett, “Ancestry-constrained phylogenetic analysis supports the indo-european steppe hypothesis,” Language, vol. 91, iss. 1, pp. 194-244, 2015.
    [Bibtex]
    @article{chang2015ancestry,
    title={Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis},
    author={Chang, Will and Cathcart, Chundra and Hall, David and Garrett, Andrew},
    journal={Language},
    volume={91},
    number={1},
    pages={194--244},
    year={2015},
    publisher={Linguistic Society of America},
    pdf={http://www.linguistics.berkeley.edu/~garrett/ChangEtAl-2015.pdf}
    }
  • [PDF] D. Hall, G. Durrett, and D. Klein, “Less grammar, more features,” in ACL, 2014.
    [Bibtex]
    @inproceedings{hall2014less,
    author = {David Hall and Greg Durrett and Dan Klein},
    title = {Less Grammar, More Features},
    year = {2014},
    booktitle = {{ACL}},
    pdf={http://www.cs.berkeley.edu/~dlwh/papers/spanparser.pdf}
    }
  • [PDF] D. Hall, T. Berg-Kirkpatrick, J. Canny, and D. Klein, “Sparser, better, faster gpu parsing,” in ACL, 2014.
    [Bibtex]
    @inproceedings{hall2014sparser,
    author = {David Hall and Taylor Berg-Kirkpatrick and John Canny and Dan Klein},
    title = {Sparser, Better, Faster GPU Parsing},
    year = {2014},
    booktitle = {{ACL}},
    pdf = {http://www.cs.berkeley.edu/~dlwh/papers/gpuparser.pdf}
    }
  • [PDF] D. Hall, A. Cohen, D. Burkett, and D. and Klein, “Faster optimal planning with partial-order pruning,” in ICAPS, 2013.
    [Bibtex]
    @inproceedings{hall2013optimal,
    author = {David Hall and Aloni Cohen and David Burkett and and Dan Klein},
    title = {Faster Optimal Planning with Partial-Order Pruning},
    year = {2013},
    booktitle = {{ICAPS}},
    pdf={http://www.aaai.org/ocs/index.php/ICAPS/ICAPS13/paper/download/6044/6168}
    }
  • [PDF] G. Durrett, D. Hall, and D. Klein, “Decentralized entity-level modeling for coreference resolution,” in ACL, 2013.
    [Bibtex]
    @InProceedings{durrett2013entity,
    author = {Durrett, Greg and Hall, David and Klein, Dan},
    title = {Decentralized Entity-Level Modeling for Coreference Resolution},
    booktitle = {{ACL}},
    year = {2013},
    url = {http://www.aclweb.org/anthology/P13-1012},
    pdf = {http://www.aclweb.org/anthology/P/P13/P13-1012.pdf}
    }
  • W. Chang, D. Hall, C. Cathcart, and A. Garrett, “Dating Proto-Indo-European: a revised computational analysis supports the steppe hypothesis,” in International Conference on Historical Linguistics, 2013.
    [Bibtex]
    @inproceedings{chang2013dating,
    author = {Will Chang and David Hall and Chundra Cathcart and Andrew Garrett },
    title = {Dating {P}roto-{I}ndo-{E}uropean: A revised computational analysis supports the steppe hypothesis},
    year = {2013},
    booktitle = {{International Conference on Historical Linguistics}}
    }
  • [PDF] J. Canny, D. Hall, and D. Klein, “A multi-teraflop constituency parser using GPUs,” in EMNLP, 2013.
    [Bibtex]
    @InProceedings{canny-hall-klein:2013:EMNLP,
    author = {Canny, John and Hall, David and Klein, Dan},
    title = {A Multi-Teraflop Constituency Parser using {GPUs}},
    booktitle = {{EMNLP}},
    year = {2013},
    url = {http://www.aclweb.org/anthology/D13-1195},
    pdf = {http://aclweb.org/anthology/D/D13/D13-1195.pdf}
    }

Projects

  • Breeze
    A linear algebra and numerics library for Scala.
  • Epic
    A high-performance parser and a framework for structured prediction problems.
  • Puck
    A GPU-accelerated parser that can create high quality constituency parses at over 400 sentences per second on even mid-range hardware.