Learning Rules: From PCFGs to Adaptor Grammars

author: Mark Johnson, Brown Laboratory for Linguistic Information Processing, Brown University

You might be experiencing some problems with Your Video player.

*Slides*
0:00	Learning Rules with Adaptor Grammars
0:53	Ideas behind talk
3:33	Questions this work tries to address
4:28	Why not study syntax?
6:50	Outline - From PCFGs to Adaptor Grammars
6:54	Probabilistic context-free grammars
7:50	A CFG for stem-suffix morphology
10:07	A “CFG” with one rule per possible morpheme
12:39	Nonparametric extensions of PCFGs
15:50	Adaptor grammars: Informal description
17:14	Adaptor grammars as generative processes
18:40	An Adaptor Grammar for stem-suffix morphology
19:13	Morphology adaptor grammar (0)
19:39	Morphology adaptor grammar (1a)
20:11	Morphology adaptor grammar (1b)
20:26	Morphology adaptor grammar (1c)
20:33	Morphology adaptor grammar (1d)
20:39	Morphology adaptor grammar (2a)
20:48	Morphology adaptor grammar (2b)
21:00	Morphology adaptor grammar (2c)
21:03	Morphology adaptor grammar (2d)
21:06	Morphology adaptor grammar (3)
21:19	Morphology adaptor grammar (4a)
21:23	Morphology adaptor grammar (4b)
21:29	Morphology adaptor grammar (4c)
21:31	Morphology adaptor grammar (4d)
21:32	Morphology as a Hierarchical Dirichlet Process
22:45	Bayesian hierarchy inverts grammatical hierarchy
23:31	Properties of adaptor grammars
24:18	Outline - Adaptor grammars for English word segmentation
24:25	Unigram adaptor grammar for English
25:39	Unigram word grammar as a Dirichlet Process
25:46	Unigram model often finds collocations
26:30	Unigram word segmentation grammar learnt
26:49	Unigram morphology adaptor grammar
28:02	Simultaneously learning word segmentation and syllable structure
28:13	Simultaneous word segmentation and syllable structure
28:27	Modeling collocations improves segmentation
29:06	Syllables + Collocations + Word segmentation
29:43	Syllables + 2-level Collocations + Word segmentation
29:45	Word segmentation results summary
32:15	Outline - Bayesian inference for Adaptor Grammars
32:22	Outline - Conclusion
32:25	Summary and future work
34:07	- Questions

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.