Advances in computational intelligence [[electronic resource] ] : theory & applications / / Fei-Yue Wang, Derong Liu

• Autore: Wang Fei-Yue
• Pubblicazione: Toh Tuck Link, Singapore ; ; New Jersey, : World Scientific, c2006
• Descrizione fisica: 1 online resource (478 p.)
• Disciplina: 006.3
• Soggetto topico: Artificial intelligence; Computational intelligence
• Soggetto genere / forma: Electronic books.
• Altri autori: LiuDerong
• Note generali: Description based upon print version of record.
• Nota di bibliografia: Includes bibliographical references and index.
• Nota di contenuto: Contents ; Preface ; List of Contributors ; 1 A Quest for Granular Computing and Logic Processing ; 1.1 Introduction ; 1.2 Granular Computing ; 1.3 Granular Computing and Logic: Synergistic Links ; 1.4 Main Categories of Fuzzy Logic Processing Units ; 1.5 A General Topology of the Network 1.6 Interpretation Issues of Logic Networks ; 1.7 Conclusions ; Bibliography ; 2 Abstraction and Linguistic Analysis of Conventional Numerical Dynamic Systems ; 2.1 Introduction ; 2.2 Type-I Linguistic Dynamic Systems ; 2.3 Type-II Linguistic Dynamic Systems 2.4 Linguistic Control Design for Goal States Specified in Words ; 2.5 Conclusions ; Bibliography ; 3 Slicing: A Distributed Learning Approach ; 3.1 Introduction ; 3.2 Slicing ; 3.3 Variance Reduction in Slicing ; 3.4 Experiments ; 3.5 Analysis 3.6 Discussion ; 3.7 Conclusions ; Bibliography ; 4 Marginal Learning Algorithms in Statistical Machine Learning ; 4.1 Introduction ; 4.2 Classification Problems and Margin ; 4.3 Maximal Margin Algorithm in SVM ; 4.4 Unbalanced Classification Problems and Weighted Maximal Margin Algorithms 4.5 The n-Unsupervised Learning Problems and Margin ; 4.6 Some Marginal Algorithms for One-Class Problems ; 4.7 Some New Algorithms of Clustering Problems ; 4.8 New Marginal Algorithms for PCA ; 4.9 Conclusions
• Sommario/riassunto: Computational Intelligence (CI) is a recently emerging area in fundamental and applied research, exploiting a number of advanced information processing technologies that mainly embody neural networks, fuzzy logic and evolutionary computation. With a major concern to exploiting the tolerance for imperfection, uncertainty, and partial truth to achieve tractability, robustness and low solution cost, it becomes evident that composing methods of CI should be working concurrently rather than separately. It is this conviction that research on the synergism of CI paradigms has experienced significant
• Titolo autorizzato: Advances in computational intelligence
• ISBN: 1-281-37912-3; 9786611379124; 981-277-392-4
• Formato: Materiale a stampa
• Livello bibliografico: Monografia
• Lingua di pubblicazione: Inglese
• Record Nr.: 9910450801503321
• Lo trovi qui: Univ. Federico II
• Serie: Series in intelligent control and intelligent automation ; ; v. 5.