ALT03Logo

The 14th International Conference
on
Algorithmic Learning Theory

Hokkaido University, Sapporo, Japan
October 17 - 19, 2003


CONFERENCE  PPROGRAM

Thursday, October 16:

17:30 - 20:00 Reception/Registration at Sapporo Aspen Hotel

(Registration desk is open at 8:30 during the conference days.)

Friday, October 17, Morning

09:00-09:10 Opening: Yuzuru Tanaka

SESSION 1, October 17, Morning

[Invited Talk] Chair: Eiji Takimoto

09:10-10:10: Thomas Zeugmann

Can Learning in the Limit be Done Efficiently?

Break (15 min)

SESSION 2, Friday, October 17, Morning

[Inductive Inference] Chair: Sanjay Jain

10:25-10:55: Sandra Zilles

Intrinsic Complexity of Uniform Learning

10:55-11:25: Eric Martin, Arun Sharma and Frank Stephan

On Ordinal VC-Dimension and Some Notions of Complexity

Short Break (10 min)

11:35-12:05: Jin Uemura and Masako Sato

Learning of Erasing Primitive Formal Systems from Positive Examples

12:05-12:35: Frank Balbach

Changing the Inference Type - Keeping the Hypothesis Space

Lunch (85 min)

SESSION 3, Friday, October 17, Afternoon

[Invited Talk] Chair: Gunter Grieser

14:00-15:00: Thomas Eiter

Abduction and the Dualization Problem

Break (30 min)

SESSION 4, Friday, October 17, Afternoon

[Learning and Information Extraction] Chair: John Case

15:30-16:00: Jan Arpe and Rüdiger Reischuk

Robust Inference of Relevant Attributes

16:00-16:30: Yusuke Suzuki, Takayoshi Shoudai, Satoshi Matsumoto, Tomoyuki Uchida and Tetsuhiro Miyahara

Efficient Learning of Ordered and Unordered Tree Patterns with Contractible Variables

Break (30 min)

SESSION 5, Friday, October 17, Afternoon

[Learning with Queries] Chair: Kouichi Hirata

17:00-17:30: Steffen Lange and Sandra Zilles

On the Learnability of Erasing Pattern Languages in the Query Model

17:30-18:00: Satoshi Matsumoto, Yusuke Suzuki, Takayoshi Shoudai, Tetsuhiro Miyahara and Tomoyuki Uchida

Learning of Finite Unions of Tree Patterns with Repeated Internal Structured Variables from Queries

SESSION 6, Saturday, October 18, Morning

[Invited Talk] Chair: Klaus P. Jantke

09:00-10:00: Genshiro Kitagawa

Signal Extraction and Knowledge Discovery Based on Statistical Modeling

Break (15 min)

SESSION 7, Saturday, October 18, Morning

[Learning with Non-linear Optimization] Chair: Jose Oncina

10:15-10:45: Jingdong Wang, Jianguo Lee and Changshui Zhang

Kernel Trick Embedded Gaussian Mixture Model

10:45-11:15: Tijl De Bie, Michinari Momma and Nello Cristianini

Efficiently Learning the Metric with Side-Information

Short Break (15 min)

11:30-12:00: Shaojun Wang and Dale Schuurmans

Learning Continuous Latent Variable Models with Bregman Divergences

12:00-12:30: Joel Ratsaby

A Stochastic Gradient Descent Algorithm for Structural Risk Minimisation

Lunch (90 min)

SESSION 8, Saturday, October 18, Afternoon

[Invited Talk] Chair: Ricard Gavaldà

14:00-15:00: Naftali Tishby

Efficient Data Representations that Preserve Information

Break (30 min)

SESSION 9, Saturday, October 18, Afternoon

[Learning from Random Examples] Chair: Hiroki Arimura

15:30-16:00: Jiri Sima

On the Complexity of Training a Single Perceptron with Programmable Synaptic Delays

16:00-16:30: John Case, Sanjay Jain, Rüdiger Reischuk, Frank Stephan and Thomas Zeugmann

Learning a Subclass of Regular Patterns in Polynomial Time

16:30-17:00: Colin de la Higuera and Jose Oncina

Identification with Probability One of Stochastic Deterministic Linear Languages

Break (60 min)

18:00-19:00: Business Meeting

19:00-21:00: Banquet (at Sapporo Aspen Hotel)

SESSION 10, Sunday, October 19, Morning

[Invited Talk] Chair: Akihiro Yamamoto

09:00-10:00: Akihiko Takano

Association Computation for Information Access

Break (15 min)

SESSION 11, Sunday, October 19, Morning

[On-Line Prediction] Chair: Frank Stephan

10:15-10:45: Ilia Nouretdinov and Vladimir Vovk

Criterion of Calibration for Transductive Confidence Machine with Limited Feedback

10:45-11:15: Vladimir Vovk

Well-Calibrated Predictions from On-Line Compression Models

Short Break (15 min)

11:30-12:00: Ilia Nouretdinov, Vladimir V'yugin and Alex Gammerman

Transductive Confidence Machine is Universal

12:00-12:30: Marcus Hutter

On the Existence and Convergence of Computable Universal Priors


We look forward to seeing you in Sapporo.