ECAI-2000 Workshop "Empirical Methods in Artificial Intelligence"

Programme

Tuesday, 22 August 2000

• On Statistically Evaluating the Quality of Empirically-Computed Hamiltonian Path
  B. John Oommen, School of Comp. Science, Carleton University, Ottawa, CANADA
  I. Kuban Altinel,Dept. of Industrial Eng. Bogazici University. Istanbul, Turkey
  Necati Aras, .I.d.e.a. A.S., Tuzla 81719, Istanbul, Turkey
  

• On Selecting Problem Instances for Evaluating Satisfiability Algorithms
  J. Marques-Silva, Instituto de Engenharia de Sistemas e Computadores (INESC), Lisboa, Portugal

• Is Davis-Putnam heavy-tailed?
  B. O'Nuallain, Institute for Language, Logic and Computation, Amsterdam, The Netherlands
  

• General discussion on analysis of algorithms (10 min)

• Focusing on the Individual - Why we Need New Methods for Characterizing Problem Difficulty
  Jean-Paul Watson and Adele E. Howe, Computer Science Department, Colorado State University, USA.

• Toward a quantification of attraction in stochastic local search
  A. Sidaner and J.-J. Chabrier, L.I.R.S.I.A., Dijon, France.
  O. Bailleux, C.R.I.L., Lens, France
  

• Probabilistic Models of Computing
  I. Rish, IBM T.J. Watson Research Center, Hawthorne, NY, USA.

• Distributed Reinforcement Learning for Multiple Objective Optimization Problems
  Carlos E. Mariano Romero, Instituto Mexicano de Tecnologia del Agua, Jiutepec, Morelos, Mexico
  Eduardo F. Morales, ITESM Campus Morelos, Morelos, Mexico
  

• Business phase classification and prediction: How to compare interpretability of classification methods?
  Claus Weihs and Ursula Sondhauss Department of Statistics, University of Dortmund, Germany
  

Information for presenters:
Each talk is assigned a 20min slot. Please leave 3-5min of this time for discussion.
For the presentation, an overhead projector and a LCD projector (beamer) will be available.