Causality Challenge #1: Causation and Prediction

WCCI08 workshop,
June 3-4, 2008, Hong-Kong Convention Center, Room 402

This workshop is part of the competition program of the competition program of the World Congress on Computational Intelligence (WCCI08).

16:50 Award ceremony
17:10 Discussion: lessons learned, planning for future events
17:30-18:30 Reception

14:00 Causality competition results -- Isabelle Guyon, Clopinet, Berkeley, California, USA
14:15 Feature Ranking Using Linear SVM -- Yin-Wen Chang, National Taiwan University
14:30 Causal and Non-causal Feature Selection for Ridge Regression -- Gavin Cawley, University of East Anglia, UK
14:45 The Partial Orientation and Local Structural Learning of DAGs for Prediction -- Jianxin Yin, Zhi Geng, M. Kuroki, Z. Cai et al., School of Mathematical Sciences, Peking University, Beijing 100871, China and Osaka University, Japan
15:00 A strategy for Making Predictions under Manipulations -- Laura Brown and Ioannis Tsamardinos, Vanderbilt Univ., US and Univ. of Crete, Greece
15:15 Random Sets Approach for Feature Selection -- Vladimir Nikulin, Suncorp, Brisbane, Australia
15:30 An Energy-based Model for Feature Selection -- Hugo Jair Escalante, Manuel Montes and Enrique Sucar, National Institute of Astrophysics, Optics and Electronics, Tonantzintla, Puebla, Mexico
15:45 Mixture Models for Markov Blanket Filtering and Classification -- Mehreen Saeed, National University of Computer and Emerging Sciences, Pakistan

Poster session
The posters will be viewed during the breaks.

Find the Relationship of Binary variable and Continuous variables -- Jinzhu Jia, School of Mathematical Sciences, Peking University,Beijing,China
Meeting the Causal Discovery Challenge with CaMML -- Julia Flores, Ann Nicholson, Owen Woodberry, Lloyd Allison, Rodney O’Donnell, Lucas R. Hope, and Kevin B. Korb, EPSA, Albacete, Spain and Monash University , Australia.
Importance of preprocessing and decision trees -- Ching-Wei Wang, University of Lincoln, UK

The competitors are encouraged to document their methods by submitting fact sheets (MSWord template, PDF template).