Hal Daumé III (NLPers)
nlpers.blogspot.com · Personal blog · on HN since 2009
Hal Daumé III, a computer science professor at the University of Maryland, writes about natural language processing and machine learning research.
What HN saysHN commenters recommend it as a good source on NLP, covering conferences and his own work, and link his post on fast baseline text categorization with Vowpal Wabbit.
25HN people shared it
8linked it in comments
19stories submitted
115story points
8 years agolast shared
2009people per year2026
Most-shared posts
- 1Google 5gram corpus has unreasonable 5grams2 people · 25 points · first shared 2010 · 9 comments on HN
- 2Fast and easy baseline text categorization with Vowpal Wabbit2 people · 5 points · first shared 2016 · discussion on HN
- 3Debugging machine learning2 people · 4 points · first shared 2016 · discussion on HN
- 4Food for thought for all researchers around2 people · 4 points · first shared 2016 · discussion on HN
- 5Whence your reward function?1 people · 35 points · first shared 2016 · 5 comments on HN
- 6Machine Learning is the new Algorithms1 people · 25 points · first shared 2014 · 2 comments on HN
- 7Machine learning is the new algorithms1 people · 3 points · first shared 2014 · discussion on HN
- 8Many opportunities for discrimination in deploying machine learning systems1 people · 3 points · first shared 2018 · discussion on HN
- 9A technique for me is a task for you1 people · 2 points · first shared 2011 · discussion on HN
- 10Teaching NLP1 people · 2 points · first shared 2012 · discussion on HN
- 11Language Bias and Black Sheep1 people · 2 points · first shared 2016 · discussion on HN
- 12Why “strong” is not a total order for NLP baselines1 people · 1 points · first shared 2015 · discussion on HN
- 13Recommended NLP, ML, and CL papers1 people · 1 points · first shared 2009 · discussion on HN
- 14A quick comment on structured input vs. structured output learning1 people · 1 points · first shared 2016 · discussion on HN
- 15My NAACL 2013 list...1 people · 1 points · first shared 2013 · discussion on HN
What HN people wrote when they linked it
ninjin · 2013-09-16
> "... it's a shame he'll likely need to get tenure before finishing it."
Judging by his homepage [1] he is at least on the tenure track. Unless the notion of an Assistant Professor is different over in the states. I agree about the sentiment about the book, at least the chapter on Neural Networks needs some serious work (I believe there are errors in there, maybe I should mail Hal). As a person and researcher I only have good things to say about him, do check out it his blog if you are into ML and NLP [2] and let's hope he will find the time to continue polishing on this freely available book (I do like his writing).
[1]: http://www.umiacs.umd.edu/~hal/
[2]: http://nlpers.blogspot.com/
abhaga · 2012-05-08
A good source of NLP goodness is http://nlpers.blogspot.in/, written by Hal Daume, a CS prof at UMD. He covers recent NLP conferences and also his own work. Some of it is in the area of domain adaptation which is of interest to anyone trying to bring research papers to real world products.
skystrife · 2017-05-22
Basically the same [1,2].
[1]: https://twitter.com/yoavgo/status/751178795323908096
[2]: https://nlpers.blogspot.com/2016/08/fast-easy-baseline-text-categoriza…
nl · 2016-08-21
Note that this is only looking at the embeddings side of FastText (which looks quite impressive).
The classification side is equally interesting, and is better compared to things like VW, eg http://nlpers.blogspot.com.au/2016/08/fast-easy-baseline-text-categori…
VW is a pretty strong baseline here, so equaling it is a good accomplishment.
BjoernKW · 2016-01-23
Though finite state machines can work for simple dialogue structures, more complex dialogues generated by them will come across as robotic.
For non-naive implementations you'll have to learn about a linguistics discipline called pragmatics, which deals with conversational semantics and context:
http://nlpers.blogspot.de/2006/04/please-pass-pragmatics.html
http://www.amazon.com/gp/product/0521294142/sr=8-1/qid=1145977367/ref=…
gtani · 2012-04-08
Good analysis of providing source code, datasets and potential burdens on reviewers and authors.
http://nlpers.blogspot.com/2011/03/some-thoughts-on-supplementary.html
alextp · 2010-08-22
A popular nlp blogger posted ( http://nlpers.blogspot.com/2010/08/readers-kill-blogs.html ) about a related problem with comments (ie, that they're getting more and more distributed).
imurray · 2010-06-24
Google released a trillion-word corpus with frequency counts for all sequences up to five words long... contains...all sorts of other errors...But the fact that it's a million times larger than the Brown Corpus outweighs these drawbacks.
Maybe not: http://nlpers.blogspot.com/2010/02/google-5gram-corpus-has-unreasonabl…