SUNDAY August 4

9:00-9:10am Introductory Remarks

9:10-11:00 Session I: Neural Nets and Related Models

9:10-9:30:

ANCHOR - A Connectionist Architecture for Hierarchical Nesting of Multiple Heterogeneous Neural Nets (Kumar and Rockett)

9:30-9:50:

A Noise-Tolerant Hybrid Model of a Global and a Local Learning Module (Oka and Yoshida)

9:50-10:00:

Break

10:00-10:20:

Combining Neural and Statistical Classifiers Via Perceptron (S. Lee)

10:20-10:50:

Handling Redundancy in Ensembles of Learned Models Using Principal Components (Merz and Pazzani)

10:50-11:00:

Questions & Answers

Not presenting:

Weight Averaging for Neural Networks and Local Resampling Schemes (Utans)

Nonlinear Combination of Neural Networks from a Diverse and Evolving Population (English)

11:00-12:30pm Lunch

12:30-2:50 Session II: Hybrids

12:30-12:50:

Combining Rules and Cases (Vanhoof and Surma)

12:50-1:10:

The Sources of Increased Accuracy for Two Proposed Boosting Algorithms (Skalak)

1:10-1:30:

Exploiting Multiple Existing Models and Learning Algorithms (Ortega)

1:30-1:40:

Break

1:40-2:10:

Generalization Capability of Homogeneous Voting Classifiers Based on Partially Replicated Data (Jelonek)

2:10-2:40:

Integrating Multiple Classifiers by Finding Their Areas of Expertise (Koppel and Engelson)

2:40-2:50:

Questions & Answers

2:50-3:15 Break

3:15-5:00 Session III: Training Efficiency

3:15-3:45:

Using Partitioning to Speed Up Specific-to-General Rule Induction (Domingos)

3:45-4:15:

A Comparative Evaluation of Combiner and Stacked Generalization (Fan, Chan and Stolfo)

4:15-4:45:

Scaling Up: Distributed Machine Learning with Cooperation (Provost and Henessey)

4:45-5:00:

Questions & Answers

MONDAY August 5

9:00-10:50am Session IV: Analytics

9:00-9:20:

Creating and Exploiting Coverage and Diversity (Brodley and Lane)

9:20-9:40:

Classifier Combining: Analytical Results and Implications (Tumer and Ghosh)

9:40-9:50:

Break

9:50-10:10:

On Voting Ensembles of Classifiers (Matan)

10:10-10:40:

On Explaining Degree of Error Reduction due to Combining Multiple Decision Trees (Ali)

10:40-10:50:

Questions & Answers

Not Presenting:

Bayesian Model Averaging: A Review (Madigan, Raftery, Volinsky and Hoeting)

10:50-12:30pm Lunch

12:30-2:50 Session V: Applications

12:30-12:50:

Selection of Learning Algorithms for Trading Systems Based on Biased Estimators (Obradovic and Chenoweth)

12:50-1:20:

On the Integration of Ensembles of Neural Networks: Application to Seismic Signal Classification (Shimshoni and Intrator)

1:20-1:30:

Break

1:30-2:00:

Human Expert-Level Performance on a Scientific Image Analysis Task by a System Using Combined Artificial Neural Networks (Cherkauer)

2:00-2:20:

Applying Parallel Learning Models of Artificial Neural Networks to Letters Recognition from Phonemes (B. Lee)

2:20-2:40:

Learning from Multiple Models to Integrate Sensed Data (Den Hartog, Elling, Mniszewski and Kieckhafer)

2:40-2:50:

Questions & Answers

2:50-3:15 Break (Topics and questions for discussion are collected)

3:15-4:45 Discussion