tag:blogger.com,1999:blog-19803222.post4754975319897954564..comments2024-03-18T01:45:45.724-06:00Comments on natural language processing blog: NAACL-HLT 2009 Retrospectivehalhttp://www.blogger.com/profile/02162908373916390369noreply@blogger.comBlogger18125tag:blogger.com,1999:blog-19803222.post-77307909273035950612010-06-08T00:39:52.738-06:002010-06-08T00:39:52.738-06:00great post.
SEO
Seo marketing
Search engine market...great post.<br /><a href="http://www.top3.com.sg/seo/" rel="nofollow">SEO</a><br /><a href="http://www.top3.com.sg/ppc/" rel="nofollow">Seo marketing</a><br /><a href="http://www.top3.com.sg/emailmarketing/" rel="nofollow">Search engine marketing</a><br /><a href="http://www.top3.com.sg/webdesign/" rel="nofollow">web design</a><br /><a href="http://www.top3.com.sg/ecommerce-solutions/" rel="nofollow">web development</a>Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-74522533543602175772010-03-02T15:34:13.556-07:002010-03-02T15:34:13.556-07:00The workshop was so good to go to. I would love to...The workshop was so good to go to. I would love to see more papers.<br /><a href="http://www.mesotheliomalawyer-columbus.com" rel="nofollow">mesothelioma lawyer columbus</a>gamefan12https://www.blogger.com/profile/17700305176595632282noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-85579677714841827732010-01-16T11:47:18.047-07:002010-01-16T11:47:18.047-07:00love to see this discussion! It’s great to see you...love to see this discussion! It’s great to see you all working through the issues and also, it’s great to see recommendations for testing. In the end, it’s what your actual users do and prefer that should be your biggest driver in making these decisions.<br /> <a href="http://www.onlineuniversalwork.com" rel="nofollow">internet work parttime</a>willsonhttps://www.blogger.com/profile/11996281370850910184noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-80650805888994937842010-01-10T00:58:07.718-07:002010-01-10T00:58:07.718-07:00I wanted to thank you for this great read!! I defi...I wanted to thank you for this great read!! I definitely enjoying every little bit of it :) I have you bookmarked to check out new stuff you post. Please visit my Please visit my blog when you have time Please come visit my site <a href="http://www.oklahomacity-classifieds.com" rel="nofollow">Free Business Listing of Reno </a> when you got time.rr8004https://www.blogger.com/profile/17668784480317088693noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-36089508519048423282010-01-10T00:57:54.835-07:002010-01-10T00:57:54.835-07:00I wanted to thank you for this great read!! I defi...I wanted to thank you for this great read!! I definitely enjoying every little bit of it :) I have you bookmarked to check out new stuff you post. Please visit my Please visit my blog when you have time Please come visit my site <a href="http://www.oklahomacity-classifieds.com" rel="nofollow">Business Services Web Directory Of Reno</a> when you got time.rr8004https://www.blogger.com/profile/17668784480317088693noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-71677275174603450982009-12-01T03:25:18.345-07:002009-12-01T03:25:18.345-07:00I just came across this article. Though I didnt at...I just came across this article. Though I didnt attend NAACL-HLT in 2009, I did read some of the papers you have written about here. Very interesting!<br /><br />Also this idea of having a retrospective blog post is a great idea. I am hoping to attend NAACL-HLT in 2010 hope to contribute more to your post then :)L. Venkata Subramaniamhttps://www.blogger.com/profile/08013877791567403788noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-45370540501263783642009-11-04T05:27:46.497-07:002009-11-04T05:27:46.497-07:00For simple logistic regression (only one layer), t...For simple logistic regression (only one layer), the fanin is simply the number of input dimensions.<br /><a rel="follow" href="http://www.writinghelp.co.uk/essay/index.asp" rel="nofollow">Essay</a> | <a rel="follow" href="http://www.writinghelp.co.uk/thesis/index.asp" rel="nofollow">Thesis</a>Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-42556664731605865262009-11-04T05:25:33.272-07:002009-11-04T05:25:33.272-07:00The vase broke" as causative/ unaccusative pa...The vase broke" as causative/ unaccusative pairs. Beth Levin wrote about the "drive" class verbs, but I don't have access to her book.<br /><a rel="follow" href="http://www.writinghelp.co.uk/assignment/index.asp" rel="nofollow">Assignment</a> | <a rel="follow" href="http://www.writinghelp.co.uk/coursework/index.asp" rel="nofollow">Coursework</a> | <a rel="follow" href="http://www.writinghelp.co.uk/dissertation/index.asp" rel="nofollow">Dissertation</a>Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-41458714416514580582009-10-30T22:29:49.199-06:002009-10-30T22:29:49.199-06:00I enjoyed reading your work! GREAT post! I looked ...I enjoyed reading your work! GREAT post! I looked around for this… but I found you! Anyway, would you mind if I threw up a backlink from my site? Please come visit my site <a href="http://www.columbusunitedstates.com" rel="nofollow">San Francisco Web Link </a> when you got time.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-87736690414246371762009-10-30T22:29:32.883-06:002009-10-30T22:29:32.883-06:00I enjoyed reading your work! GREAT post! I looked ...I enjoyed reading your work! GREAT post! I looked around for this… but I found you! Anyway, would you mind if I threw up a backlink from my site? Please come visit my site <a href="http://www.columbusunitedstates.com" rel="nofollow">San Francisco Business Search Engine</a> when you got time.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-79688655837216153612009-10-24T01:33:07.958-06:002009-10-24T01:33:07.958-06:00I really liked your blog! Please come visit my sit...I really liked your blog! Please come visit my site <a href="http://www.businessesnashville.com" rel="nofollow">Business Directory Nashville </a> when you got time.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-89993770925893719922009-10-24T01:32:34.337-06:002009-10-24T01:32:34.337-06:00I really liked your blog! Please come visit my sit...I really liked your blog! Please come visit my site <a href="http://www.businessesnashville.com" rel="nofollow">Phone Directory Of Nashville City Tennessee TN State </a> when you got time.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-75752616797033650152009-10-15T00:47:03.613-06:002009-10-15T00:47:03.613-06:00and your success is very much an inspiration for m...and your success is very much an inspiration for me. Please come visit my site <a href="http://www.mosaintlouis.com" rel="nofollow">Business Services Web Directory Of St. Louis</a> when you got time.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-64649907162263796342009-10-15T00:46:30.478-06:002009-10-15T00:46:30.478-06:00and your success is very much an inspiration for m...and your success is very much an inspiration for me. Please come visit my site <a href="http://www.kansascityprofile.com" rel="nofollow">Kansas City City Business Search Engine</a> when you got time.Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-19803222.post-10947633603876627112009-06-15T20:18:31.956-06:002009-06-15T20:18:31.956-06:00Alan Ramsey gave a good paper at the end of the Li...Alan Ramsey gave a good paper at the end of the Linguistic Creativity workshop. It was called "Sorry seems to be the hardest word", and consisted of a series of examples of things that "I am sorry that X" could mean. This has the golden property that it draws attention to facts about language that are a challenge to both the stats-heavy and the logic-heavy subgroups of ACL. There seems to be a need for common-sense reasoning here. The temptation is to dismiss this as hopelessly retro and do something else instead. That would be wrong.Chris Brewhttps://www.blogger.com/profile/15950294272852443488noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-27660633903286855212009-06-15T08:10:29.767-06:002009-06-15T08:10:29.767-06:00Oops, yes... it's definitely more like a causa...Oops, yes... it's definitely more like a causative than a passive. Thanks for catching the error.<br /><br />BTW, I also realized I totally missed the best papers, both of which were really great. I'll edit the post in a few minutes.halhttps://www.blogger.com/profile/02162908373916390369noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-27014570928409176852009-06-14T10:58:11.700-06:002009-06-14T10:58:11.700-06:00Re #8, you write "passivization in English, w...Re #8, you write "passivization in English, where you might say something like 'The truck drove into the crowd' rather than 'The soldier drove the truck into the crowd.'" Actually, the passive here would be "The truck was driven into the crowd." I'm not sure there's a name for the sentence "The truck drove into the crowd"; it's similar in some ways to what was called a Middle Voice in Greek. Pustejovsky (who BTW gave a poster at NAACL) referred to pairs like "John broke the vase" and "The vase broke" as causative/ unaccusative pairs. Beth Levin wrote about the "drive" class verbs, but I don't have access to her book.Unknownhttps://www.blogger.com/profile/13894825668610190284noreply@blogger.comtag:blogger.com,1999:blog-19803222.post-91092441867141209882009-06-12T22:42:57.805-06:002009-06-12T22:42:57.805-06:00Dan Klein was also asking about how to initialize ...Dan Klein was also asking about how to initialize the weights with the quadratic filters setup. Here is the technique I used for chunking. (James Bergstra has a slightly different setup for his vision experiments.)<br /><br />For a particular layer of training weights, let the fanin be the number of units in the input. For simple logistic regression (only one layer), the fanin is simply the number of input dimensions.<br /><br />Let v = 1/sqrt(fanin)<br /><br />We initialize the bias terms to be 0.<br />We initialize the weights uniformly in the range [-v, +v].<br />We initialize the quadratic filters uniformly in the range [-v*qscale, +v*qscale]. qscale = 0.01 worked fine for us.<br /><br />For future reference, when doing SGD training it helps to scale the<br />learning rate by v. This only makes a difference if there is more than<br />one layer (e.g. single hidden layer net).<br /><br />See you chez nous, at ICML!Joseph Turianhttps://www.blogger.com/profile/06249878639857416906noreply@blogger.com