Information, Statistics and Induction in Science

Proceedings of the Conference, ISIS '96

Conference, ISIS '96

, Melbourne, Australia

, 20 – 23 August 1996

https://doi.org/10.1142/3251 | July 1996

Pages: 420

Edited by:
David L Dowe (Monash University, Australia),
Kevin B Korb (Monash University, Australia), and
Jonathan J Oliver (Monash University, Australia)

View Full Book

Tools

Purchase Save for later

ISBN: 978-981-4547-26-0 (ebook) USD 47.00 Add to cart

This conference will explore the use of computational modelling to understand and emulate inductive processes in science. The problems involved in building and using such computer models reflect methodological and foundational concerns common to a variety of academic disciplines, especially statistics, artificial intelligence (AI) and the philosophy of science. This conference aims to bring together researchers from these and related fields to present new computational technologies for supporting or analysing scientific inference and to engage in collegial debate over the merits and difficulties underlying the various approaches to automating inductive and statistical inference.

The proceedings also include abstracts by the invited speakers (J R Quinlan, J J Rissanen, M Minsky, R J Solomonoff & H Kyburg, Jr.).

Sample Chapter(s)
First-Order Induction: Techniques and Applications (71 KB)

Contents:

Use of Randomization to Normalize Feature Merits (S J Hong et al.)
Constructive Induction of Cartesian Product Attributes (M J Pazzani)
Minimum Information Estimation of Linear Regression Models (S Legg)
Maximum Expected Utility Principle: The Case Study of Information Retrieval (G Amati et al.)
Induction in Medical Discovery: A Computational Simulation (V Corruble)
Measuring Randomness by Rissanen's Stochastic Complexity: Applications to the Financial Data (S H Chen & C W Tan)
Computer Based Life, Possibilities and Impossibilities (A Dorin)
From Statistical Regularities to Concepts, Hierarchies, Equation Clusters and Rules (J M Zytkow)
Ideal MDL and Its Relation to Bayesianism (P Vitányi & M Li)
False Oracles and Strict MML Estimators (C S Wallace)
An Algorithm for Unsupervised Learning via Normal Mixture Models (G J McLachlan & D Peel)
A Minimum Message Length (MML) Model for Software Measures (J Patrick)
and other papers

Readership: Computer scientists, statisticians and philosophers.

No Access

Information, Statistics and Induction in Science: Proceedings of the Conference, ISIS '96

Pages:1–423

https://doi.org/10.1142/9789814530637

Preview Abstract

The Table of Contents for the book is as follows:

Tutorials
Invited Presentations
- First-Order Induction: Techniques and Applications
- A Universal Regression Model
- Discovering Theories of Discovering
- Bayesian Inference and Inductive Inference
- Does Algorithmic Probability Solve the Problem of Induction?
Classification Trees, Graphs and Rules
- Use of Randomization to Normalize Feature Merits
- A Heuristic Covering Algorithm has Higher Predictive Accuracy Than Learning All Rules
- Single Pass Constructive Induction with Continuous Variables
- An MDL Estimate of the Significance of Rules
- Producing More Comprehensible Models While Retaining Their Performance
- Constructive Induction of Cartesian Product Attributes
- The MML Evolution of Classification Graphs
Regression
- Estimation of Regression Disturbances Based on Minimum Message Length
- Minimum Information Estimation of Linear Regression Models
Foundations of Statistics
- The Evaluation of Model Selection Criteria: Pointwise Limits in the Parameter Space
- Fuzzy Hypothesis Tests and Confidence Intervals
- Maximum Expected Utility Principle: The Case Study of Information Retrieval
- Multi-Layer Thinking in Logic and Probability
Biology
- Fuzzy Gating and Its Application in Flow Cytometry
- Induction in Medical Discovery: A Computational Simulation
- Comparative Analysis of Amino Acid Sequences of Proteins Using Rough Sets and Change of Representation
- Minimum Complexity Principle and its Application to Reconstruction of Molecular Phylogenetic Tree
Economics
- Measuring Randomness by Rissanen's Stochastic Complexity: Applications to the Financial Data
- Conceptual Difficulties with the Efficient Market Hypothesis: Towards a Naturalized Economics
Philosophy of AI
- Chance Lowering Causes: Old Problems for New Versions of the Probabilistic Theory of Causation
- Computer Based Life, Possibilities and Impossibilities
- Symbolicism and Connectionism: AI Back at a Join Point
Scientific Discovery
- Inductive Theories from Equational Systems
- From Statistical Regularities to Concepts, Hierarchies, Equation Clusters and Rules
Minimum Encoding Inference
- Ideal MDL and Its Relation to Bayesianism
- The Likelihood Principle and MML Estimators
- False Oracles and Strict MML Estimators
- An Analysis of SMML from a Subjective Bayesian Perspective
Mixture Modeling
- Estimating the Number of Components in a Normal Mixture
- Comparison of Unsupervised Classifiers
- An Algorithm for Unsupervised Learning via Normal Mixture Models
- Comparing Bayesian Model Class Selection Criteria by Discrete Finite Mixtures
- Mixture Model Clustering of Data Sets with Categorical and Continuous Variables
- A Minimum Message Length (MML) Model for Software Measures
Author Index