Induction as optimization
IEEE Transactions on Systems Man and CyberneticsPublished 1 January 1990
Larry Rendell
Citations12
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
It is concluded that combinations of strengths can overcome difficulties in domains that are simultaneously probabilistic, dynamic, noisy, and biased.
Abstract
A representative (but unimodal) problem, the fifteen puzzle, is used to test six different learning systems: some that fit, count, or partition data in instance space; some that optimize measures derived from data in hypothesis space; and some that perform combinations of such procedures. The six systems are described, tested, and analyzed. Through several experiments, specific properties are extracted. By combining two or three kinds of techniques, the extent to which they complement each other is gauged. It is concluded that combinations of strengths can overcome difficulties in domains that are simultaneously probabilistic, dynamic, noisy, and biased.>
Keywords
Computer Science
The MIT Press eBooksAdaptation in Natural and Artificial Systems
35,568 Citations1992John H. Holland
Initially applying his concepts to simply defined artificial systems with limited numbers of parameters, Holland goes on to explore their use in the study of a wide range of complex, naturally occuring processes, concentrating on systems having multiple factors that interact in nonlinear ways.
BiometricsClassification and Regression Trees.
23,841 Citations1984Alexander Gordon, Leo Breiman +3 more
Cambridge University Press eBooksJudgment under Uncertainty
8,011 Citations1982Kahneman, Daniel 1934-2024, Slovic, Paul 1938- +1 more
IBM Journal of Research and DevelopmentSome Studies in Machine Learning Using the Game of Checkers
4,355 Citations1959Arthur L. Samuel
A new signature-table technique is described together with an improved book-learning procedure which is thought to be much superior to the linear polynomial method and to permit the program to look ahead to a much greater depth than it otherwise could do.
Journal of the Royal Statistical Society Series A (General)Pattern Classification and Scene Analysis.
3,240 Citations1974Mairi Clarke, Richard O. Duda +1 more
Pattern Recognition Principles
3,205 Citations2009
The present work gives an account of basic principles and available techniques for the analysis and design of pattern processing and recognition systems.
Machine LearningExplanation-Based Generalization: A Unifying View
1,175 Citations1986Tom M. Mitchell, Richard M. Keller +1 more
This paper proposes a general, domain-independent mechanism, called EBG, that unifies previous approaches to explanation-based generalization, and is illustrated in the context of several example problems, and used to contrast several existing systems for explanation- based generalization.
Machine LearningLearning Efficient Classification Procedures and Their Application to Chess End Games
1,106 Citations1983J. R. Quinlan
A series of experiments dealing with the discovery of efficient classification procedures from large numbers of examples is described, with a case study from the chess end game king-rook versus king-knight.
Behavioral and Brain SciencesOn artificial intelligence
885 Citations1985Peter H. Schönemann
This paper aims to boost the accuracy of a DP logistic regression (LR) via a pre-training module via a pre-training module and shows that adding a pre-training module significantly improves the accuracy of the DP-LR model.
Machine LearningExplanation-based generalization: A unifying view
718 Citations1986Tom M. Mitchell, Richard M. Keller +1 more
This paper proposes a general, domain-independent mechanism, called EBG, that unifies previous approaches to explanation-based generalization, and is illustrated in the context of several example problems, and used to contrast several existing systems for explanation- based generalization.
The MIT Press eBooksProduction System Models of Learning and Development
496 Citations1987
This is the first book to focus exclusively on these important models of human cognition, collecting and giving many of the best examples of current research.
Shift of bias for inductive concept learning
288 Citations1984Paul E. Utgoff
It is shown that search for appropriate bias is itself a major part of the learning task, and that mechanical procedures for conducting a well-directed search for an appropriate bias can be created.
Cognitive ScienceModels of Concepts*
252 Citations1984Benjamin Cohen, Gregory L. Murphy
A knowledge representation model of prototype theory is outlined, based on work in schema theory and AI knowledge representation, and it is argued that if this model is model concepts as knowledge representations of a certain kind, it is possible to answer prototype theory's critics, but to address more fundamental issues in the theory of concepts.
International Joint Conference on Artificial IntelligenceIn defense of probability
238 Citations1985Peter Cheeseman
Proceedings of the Royal Society of London A Mathematical and Physical SciencesExperiments with the Graph Traverser program
205 Citations1966J. Doran, Donald Michie
An automatic method is described for the solution of a certain family of problems that depends upon the evaluation of intermediate states of the problem according to the extent to which they have features in common with the goal state.
Some Studies in Machine Learning Using the Game of Checkers. I
148 Citations1988Arthur L. Samuel
Machine LearningA General Framework for Induction and a Study of Selective Induction
135 Citations1986Larry Rendell
This paper offers a synthetic view of induction which exploits uncertainty, task-guidance, and biases such as language restriction, and suggests a number of incipient principles for powerful induction.
Elsevier eBooksRecent Results on Boolean Concept Learning
130 Citations1987Michael Kearns, Ming Li +2 more
A recently introduced model of concept learning that pays particular attention to computational feasibility is described and the various directions in which the model can be extended are discussed.
Artificial IntelligenceA new basis for state-space learning systems and a successful implementation
99 Citations1983Larry Rendell
A new basis for state-space learning systems is described which centers on a performance measure localized in feature space, and despite the absence of any objective function the parameter vector is locally optimal.
National Conference on Artificial IntelligenceA framework for comparing alternative formalisms for plausible reasoning
99 Citations1986Eric Horvitz, David Heckerman +1 more
It is demonstrated that the logical relationship can facilitate the identification of differences among alternative plausible reasoning methodologies and make use of the relationship to examine popular non-probabilistic strategies.
Computational IntelligenceLearning hard concepts through constructive induction: framework and rationale
97 Citations1990Larry Rendell, Raj Seshu
This work argues for a specific approach to constructive induction that reduces variation by incorporating various kinds of domain knowledge, i.e., transformations that group together non‐contiguous portions of feature space having similar class‐membership values.
Elsevier eBooksAn Empirical Comparison of Genetic and Decision-Tree Classifiers
95 Citations1988J. R. Quinlan
Results obtained by a genetic learning system Boole on a small, abstract learning task are reported and conversion of the decision trees to sets of production rules has led to a significant improvement in classification accuracy for this task.
Machine LearningA general framework for induction and a study of selective induction
82 Citations1986Larry Rendell
Layered concept-learning and dynamically variable bias management
70 Citations1987Larry Rendell, Raj Sheshu +1 more
A model of concept formation is presented that views learning as a simultaneous optimization problem at three different levels, with dynamically chosen biases guiding the search for satisfactory hypotheses.
Genetic Plans and the Probabilistic Learning System: Synthesis and Results
28 Citations1985Larry Rendell
This paper describes new conceptual and experi mental results using the probabilistic learning system PLSe, PLS2, designed for any task in which overall performance can be measured, and in which choice or task objects or operators influences perrormance.
The effect of data character on empirical concept learning
10 Citations2003Larry Rendell, H. Cho
The authors examine typical problems in an effort to characterize concepts more completely and utilize the characterization to generate data artificially to measure learning performance as a precise function of several data characteristics.
An adaptive plan for state-space problems
7 Citations1981Larry Rendell
An adaptive plan (meta-strategy) is described which automatically generates evaluation functions for state-space problems, and which has created a function that solves the fifteen puzzle with (locally) optimal parameters.