| DATE | AGENDA |
|---|
| 2005/09/08 |
Preliminary meeting to discuss future reading-group
meetings. |
| 2005/09/15 |
Igor will present Hidden-Variable Models for Discriminative Reranking by Koo and Collins.
Much fruitful discussion will ensue.
|
| 2005/09/22 |
Kevin will present
Learning
to Map sentences to Logical Form: Structured Classification
with Probabilistic Categorial Grammars by
Zettlemoyer and Collins
|
| 2005/09/29 |
Yoong Keok and Kevin will discuss one of their research problems (learning a global partitioning function for coreference resolution) and some related papers ("Learning Spectral Clustering" by Bach and Jordan; "Regularized spectral learning UW-Stat Dept TR # 465" by Meila, Shortreed, and Xu; and "Pairwise Clustering and Graphical Models" by Shental, Hertz, Zomet, and Weiss).
|
| 2005/10/06 |
Regina will be at EMNLP. No reading group unless by popular demand.
|
| 2005/10/11 |
As Thursday is Yom Kippur, reading group will be held on TUESDAY this week.
Ben will present
Contrastive Estimation: Training Log-Linear Models on Unlabeled Data
by Smith and Eisner.
|
| 2005/10/20 |
Yoong Keok will present
A High-Performance Semi-Supervised Learning Method for Text Chunking
by Ando and Zhang. A related paper, which you may wish to consult, is
Bootstrapping Without the Boot by Eisner and Karakos (EMNLP 2005).
|
| 2005/10/27 |
No reading group this week due to a talk by Chris Manning.
|
| 2005/11/10 | Philip will present
Incorporating Non-local Information into Information
Extraction with Gibbs Sampling
by
Jenny Rose Finkel, Trond Grenager and Christopher Manning.
|
| 2005/11/17 | Ben will present
A Discriminative Matching Approach to Word Alignment
by Dan Klein, Simon Lacoste-Julien, and Ben Taskar.
He promises a rigorous treatment, so come prepared.
|
| 2005/11/21 |
Igor will present his research on segmentation
on Monday at 1pm (32-G451). No reading required.
|
| 2005/12/08 |
Kevin will present
Seeing Stars: Exploiting Class Relationships for Sentiment Categorization with Respect to Rating Scales
by Pang and Lee;
|