Selecting Models from Data

1. Front Matter

   Pages i-x

   PDF

2. Overviews: Model Selection

   1. Front Matter

      Pages 1-1

      PDF

   2. Statistical strategy: step 1

      • D. J. Hand

      Pages 3-9

   3. Rational Learning: Finding a Balance Between Utility and Efficiency

      • Jonathan Gratch, Gerald DeJong, Yuhong Yang

      Pages 11-20

   4. A new criterion for selecting models from partially observed data

      • Hidetoshi Shimodaira

      Pages 21-29

   5. Small-sample and large-sample statistical model selection criteria

      • S. L. Sclove

      Pages 31-39

   6. On the choice of penalty term in generalized FPE criterion

      • Ping Zhang

      Pages 41-49

   7. Cross-Validation, Stacking and Bi-Level Stacking: Meta-Methods for Classification Learning

      • Cullen Schaffer

      Pages 51-59

   8. Probabilistic approach to model selection: comparison with unstructured data set

      • Victor L. Brailovsky

      Pages 61-70

   9. Detecting and Explaining Dependencies in Execution Traces

      • Adele E. Howe, Paul R. Cohen

      Pages 71-77

   10. A method for the dynamic selection of models under time constraints

       • Geoffrey Rutledge, Ross Shachter

       Pages 79-88

3. Graphical Methods

   1. Front Matter

      Pages 89-89

      PDF

   2. Strategies for Graphical Model Selection

      • David Madigan, Adrian E. Raftery, Jeremy C. York, Jeffrey M. Bradshaw, Russell G. Almond

      Pages 91-100

   3. Conditional dependence in probabilistic networks

      • Remco R. Bouckaert

      Pages 101-111

   4. Reuse and sharing of graphical belief network components

      • Russell Almond, Jeffrey Bradshaw, David Madigan

      Pages 113-122

   5. Bayesian Graphical Models for Predicting Errors in Databases

      • David Madigan, Jeremy C. York, Jeffrey M. Bradshaw, Russell G. Almond

      Pages 123-131

   6. Model Selection for Diagnosis and Treatment Using Temporal Influence Diagrams

      • Gregory M. Provan

      Pages 133-142

   7. Diagnostic systems by model selection: a case study

      • S. L. Lauritzen, B. Thiesson, D. J. Spiegelhalter

      Pages 143-152

   8. A Survey of Sampling Methods for Inference on Directed Graphs

      • Andrew Runnalls

      Pages 153-162

   9. Minimizing decision table sizes in influence diagrams: dimension shrinking

      • Nevin Lianwen Zhang, Runping Qi, David Poole

      Pages 163-172

This volume is a selection of papers presented at the Fourth International Workshop on Artificial Intelligence and Statistics held in January 1993. These biennial workshops have succeeded in bringing together researchers from Artificial Intelligence and from Statistics to discuss problems of mutual interest. The exchange has broadened research in both fields and has strongly encour­ aged interdisciplinary work. The theme ofthe 1993 AI and Statistics workshop was: "Selecting Models from Data". The papers in this volume attest to the diversity of approaches to model selection and to the ubiquity of the problem. Both statistics and artificial intelligence have independently developed approaches to model selection and the corresponding algorithms to implement them. But as these papers make clear, there is a high degree of overlap between the different approaches. In particular, there is agreement that the fundamental problem is the avoidence of "overfitting"-Le., where a model fits the given data very closely, but is a poor predictor for new data; in other words, the model has partly fitted the "noise" in the original data.

• Hidden Markov Model
• artificial intelligence
• autonomous agents
• best fit
• calculus
• databases
• decision tree
• expert system
• knowledge
• knowledge-based system
• learning
• modeling
• neural network
• statistical model
• statistics

• Ames Research Center, NASA, Moffet Field, USA

  P. Cheeseman

• Department of Statistics and Actuarial Science, University of Waterloo, Waterloo Ontario, Canada

  R. W. Oldford

Policies and ethics