Reading Group on The Elements of Statistical Learning
Language and Inference Technology
Group
ILLC, University of Amsterdam
Nieuwe Achtergracht 166
1018 WV Amsterdam, The Netherlands
Time: Monday 10:0-11:00
Room: B235
Overview: We are interested in statistical-learning methods
(such as nearest-neighbor methods, bootstrap and maximum-likelihood
methods, boosting, neural networks, support-vector machines,
co-training, and maximum-entropy modeling), and we are especially
interested in application of these methods to Natural-Language
Processing (NLP). So, we plan to read some selected chapters of
which will provide us with the theoretical background of these
methods. Additionally, we plan to discuss papers applying these
methods to (tasks involving at least some) NLP.
Reading Group Members (so far):
Gabriel Infante Lopez (infante@science.uva.nl),
Valentin Jijkoun (jijkoun@science.uva.nl),
Karin Müller (kmueller@science.uva.nl),
Breanndan O Nuallain (bon@science.uva.nl),
Detlef Prescher (prescher@science.uva.nl),
Yoav Seginer (yseginer@science.uva.nl), and
Khalil Sima'an (simaan@science.uva.nl).
Syllabus (so far):
Introduction
[Session chair (TUESDAY, February 25, 2003): Khalil Sima'an]
Overview of Supervised Learning, Chapter 2 (Hastie etal.,
2001)
[Session chair (March 03, 2003): Detlef Prescher]
Bayesian Decision Theory, Chapter 2 (Duda etal., 2001), and
Wray L. Buntine
(1994). Operations for Learning with Graphical Models..
[Session chair (March 10, 2003): Valentin Jijkoun]
[Session chair (March 17, 2003): Yoav Seginer]
Maximum-Likelihood and Bayesian Parameter Estimation, Chapter 3
(Duda etal., 2001).
[Session chair (March 24, 2003): Breanndan O Nuallain]
[Session chair (THURSDAY, April 3, 2003): Breanndan O Nuallain]
Nonparametric Techniques, Chapter 4 (Duda etal., 2001).
[Session chair (May 12, 2003): Yoav Seginer]
Linear Discriminant Functions, Chapter 5 (Duda etal., 2001).
[Session chair (May 19, 2003): Gabriel Infante Lopez]
[Session chair (May 26, 2003): Valentin Jijkoun]
[Session chair (June 2, 2003): Valentin Jijkoun]
Support Vector Machines, Chapter 12 (Hastie etal., 2001), and
Christopher Burges(1998).
Tutorial on Support-Vector Machines for Pattern Recognition, and
Thorsten
Joachims (2001).
A Statistical Learning Model of Text
Classification with Support Vector Machines, and
Collins
(2002). Parameter Estimation for Statistical Parsing Models:
Theory and Practice of Distribution-Free Methods, and
Collins and Duffy (2002). New Ranking Algorithms for Parsing and
Tagging: Kernels over Discrete Structures, and the Voted
Perceptron.
- [Session chair (June 16, 2003): Detlef Prescher]
Model Assesment and Selection. Chapter 7 (Hastie etal.,
2001).
[Session chair: Khalil Sima'an]
Model Inference and Averaging, Chapter 8 (Hastie etal.,
2001) and
Steven Abney
(2002). Bootstrapping.
[Session chair: Detlef Prescher]
Boosting, Chapter 10 (Hastie etal., 2001), and
Henderson
and Brill (2000). Bagging and Boosting a Treebank
Parser.
[Session chair: Gabriel Infante Lopez]
Neural Networks, Chapter 11 (Hastie etal., 2001)
[Session chair: Valentin Jijkoun]
Co-Training,
Blum and Mitchell (1998). Combining Labeled and Unlabeled Data
with Co-Training.
[Session chair:]
Maximum-Entropy Modeling,
Robert Malouf
(2002). A comparison of algorithms for maximum entropy parameter
estimation, and
Adwait
Ratnaparkhi (1997). A Simple Introduction to Maximum Entropy
Models for Natural Language Processing.
[Session chair:]
Last updated: June 2003.