Compendium of Machine Learning I
Symbolic Learning
Häftad, Engelska, 1996
379 kr
Tillfälligt slut
Machine learning is a relatively new branch of artificial intelligence. The field has undergone a significant period of growth in the 1990s, with many new areas of research and development being explored. In this text, the authors take a broad brush view of the field in an attempt to bring together many of the results that have been achieved, presenting a general taxonomy of the field and the various key representative algorithms.
Produktinformation
- Utgivningsdatum1996-05-01
- FormatHäftad
- SpråkEngelska
- Antal sidor200
- FörlagIntellect
- ISBN9781567501780
Tillhör följande kategorier
- Introduction1 Definitions, Paradigms, Taxonomies1.1 What Is Machine Learning?1.2 Paradigms1.3 Taxonomies1.4 Representation of Acquired Concepts1.5 Background Knowledge1.6 Comparison of Techniques1.7 Knowledge-Level vs. Symbol-Level1.8 Theoretical and Empirical EvaluationSymbolic Empirical Learning2 Introduction to SEL3 Learning From Examples3.1 Description Languages3.2 Learning As Search3.3 Single vs. Multiple-concept Learning3.4 Incremental vs. Batch Learning3.5 The Importance of Inductive Bias3.6 The Single Representation Trick3.7 The Need for Constructive Induction3.8 The Problem of Noisy Data3.9 Source of Instances4 Decision Trees4.1 Decision Trees as Concept Classifiers4.2 Representational Restrictions4.3 The TDIDT Family Tree4.4 Evaluation of the TDIDT Method4.5 CLs-Concept Learning System4.5.1 General CLS Algorithm4.6 ID34.6.1 Windowing4.6.2 Problems with ID34.6.3 Noise, Missing Values, and Pruning4.7 Related Systems and Recent Work4.7.1 ACLS4. 7.2 ASSISTANT4.7.3 C4 , C4.54.7.4 CART4.7.5 FRINGE4.7.6 M54.7.7 MARS4.7.8 PLSl4.7.9 Conditional Rule Generation (CRG)4.7.10 Decision Graphs4.8 Alternative Test Selection Heuristics4.9 Inclusion of Background Knowledge4.10 Discovery of New Features4.11 Incremental Processing of Examples4.12 Continuous-Valued Attributes5 Version Spaces5.1 Basic Version Space Algorithm5.2 Discussion of the Version Space Method5.3 Representational Restrictions6 Covering Algorithms6.1 The AQ Star Methodology6.1.1 Simplified Star Algorithm6.1.2 Problem Background Knowledge6.1.3 Generalization Rules6.2 AQll6.2.1 AQ156.2.2 AQTT-15 and POSEIDON6.3 INDUCE6.3.1 The INDUCE Algorithm6.4 RIGEL6.5 Discussion of AQ-Based Methods6.6 Least Generalization 6.6.1 Plotkin6.6.2 Algorithm for Least Generalization6.7 DLG6.7.1 The DLG Algorithm6.7.2 Discussion of DLG6.8 Other Least Generalization Systems6.9 Other Covering Systems6.9.1 CN 26.9.2 Decision Lists6.10 Clustering and Numerical Systems7 Inductive Logic Programming7.1 FOIL7.1.1 The FOIL Algorithm7.1.2 Limitations and Discussion7.2 GOLEM7.2.1 The GoLEM Algorithm.7.3 Other Recent ILP Systems8 Inductive Bias9 Conceptual Clustering9.1 CLUSTER/29.1.1 The CLUSTER/2 Algorithm9.1.2 CLUSTER/S9.2 COBWEB9.2.1 Category Utility9.2.2 Representation of Concepts9.2.3 Operators 9.2.4 The COBWEB Algorithm9.2.5 Discussion of COBWEB9.2.6 Related Systems9.3 UNIMEM9. 3.1 The UNIMEM Algorithm9.3.2 RE SEARCHER9.4 WITT9.5 Other Conceptual Clustering Systems10 Machine Discovery10.1 AM10.1.1 The Architecture of AM10.1.2 Discussion of AM10.2 EURISKO10.3 BACON10.3.1 Summary of the BACON Programs10.3.2 Detecting Trends and Constants10.3.3 BACON'S Rule-Space Operators 10.3.4 Intrinsic Properties and Common Divisors10.3.5 Discussion of the BACON Method 10.3.6 Related Discovery Systems 10.4 ABACUS 10.5 PHINEAS10.6 Other Discovery SystemsAppendix: Other SEL TopicsAnalytical Learning11 Introduction to EBL11.1 EBL and Human Learning11.2 Bias and Domain Knowledge11.3 Imperfect Domain Theory11.4 The Utility Problem11.5 Operationality 11.6 Operationality and Generality 11.7 Representations and Learning 12 Composite Rules12.1 EEG-Explanation-Based Generalization12.1.1 The EBG Algorithm12.1.2 MEBG-Multiple Example EBG12.2 EGGS12.2.1 The EGGS Algorithm12.3 GENESIS12.4 BAGGER 2 12.5 Equivalence of Algorithms12.6 Other Macro-Operator Systems13 Search Control Knowledge13.1 LEX213.1.1 METALEX13.2 PRODIGY13.3 SOAR13.4 Other Search Control SystemsAppendix: Other EBL TopicsExemplars, Case-Based Reasoning, and Analogy14 Exemplar-Based Learning14.1 IBL14.1.1 The lBL Algorithms14.1.2 Similarity Function14.2 PROTOS14.2.1 PROTOS Classification Algorithm15 Case-Based Reasoning15,l JUDGE15.2 CHEF Appendix: Other Exemplar, Case-Based Topics16 Learning by Analogy16.1 Diagrammatic View16.2 The Analogy Process16.3 Modes of Analogy16.3.1 Proportional Analogy16,3.2 Predictive Analogy16,3.3 Interpretive Analogy16.4 COPYCAT16.5 ANALOGY16.6 Derivational Analogy16. 7 Structure Mapping Theory16.8 PUPS16.9 Purpose-Directed AnalogyAppendix: Other Analogy TopicsIntegrated Learning Systems17 Introduction to Integrated Systems18 Overly General or Overly Specific Theories18.1 IOE 18.1.1 Semantic Bias 18.1.2 Discussion of the Method 18.1.3 Vapnik-Chervonenkis Dimension18.2 Jou18,2.1 The Jou Algorithm18.3 Incremental Version Space Merging18.3.1 The IVSM Algorithm18.3.2 An Example of the IVSM Method18.3.3 Discussion of the IVSM Method18.4 Other Systems for Overly General Theories18.5 Overly Specific Domain Theories18.6 Learning by Failing to Explain18.7 SIERRA18.8 Other Systems for Overly Specific Theories19 Systems for General Theory Revision19.1 ML-SMART19.1.1 The ML-SMART Algorithm19.1.2 Discussion of the Method19.2 FoCL19.3 EITHER19.3.1 An Example of EITHER19.3.2 Theory for Data Interpretation19.3.3 Discussion of the Method19.4 FORTE19.4.1 Inverse Resolution19.5 OCCAM19.6 Other Systems for Theory Revision19.7 Abduction19.8 Leaming Apprentice Systems19.8.1 LEAP19.8.2 DISCIPLE19.8.3 ODYSSEUS19.8.4 CLINT-CIA19.9 Knowledge Acquisition SystemsAppendix: Other Integrated System TopicsFormal Analysis-Theory20 Machine Learning Theory20.1 Gold20.2 Valiant20.3 Blumer Bound20.4 Bias20.5 DeMorgan's Rules20.6 Valiant's Algorithm fork-CNF20.7 Vapnik-Chervonenkis Dimension20.8 Example PAC Analysis20.9 Structural Domains and Leamability20.1 0 Average-Case AnalysisAppendix: Other Formal T heory TopicsAppendicesA Glossary