Models of language acquisition : inductive and deductive approaches
Author(s)
Bibliographic Information
Models of language acquisition : inductive and deductive approaches
(Oxford linguistics)
Oxford University Press, 2002, c2000
- : pbk
Available at / 31 libraries
-
No Libraries matched.
- Remove all filters.
Note
2003 reprinted and 2004 printing has only copyright year
Includes bibliographical references and index
Series statement from cover
Description and Table of Contents
- Volume
-
ISBN 9780198299899
Description
This book presents recent advances by leading researchers in computational modelling of language acquisition. Sophisticated theoretical models can now be tested using simulation techniques and large corpora of linguistic data. Renewed interest in learning neural networks and the ability to test new solutions to fundamental problems has fuelled debates in an already very active field. The twenty-four authors in this collection of new work have been drawn from
departments of linguistics, cognitive science, psychology, and computer science. The book as a whole shows what light may be thrown on fundamental problems when powerful computational techniques are combined with real data
A central question addressed in the book concerns the extent to which linguistic structure is readily available in the environment. The authors consider the evidence in relation to word boundaries and phonotactic structure, stress patterns, text-to-speech rules, and the mapping of lexical semantics, one author arguing that a childs own output may serve as a key source of linguistic input. Linguistic structure-environment relations are central to the debate on the degree to which language
learning is inductive or deductive: this issue is considered here in studies of the acquisition of pluralization and inflectional morphology.
The book examines the power and utility of different modeling formalisms for different problems and approaches: how far, for example, can connectionist models be used as models for language acquisition or Simple Recurrent Networks form the basis of a model of language acquisition? To what degree can lexical items and categories be used in the construction of neural network models, or Markov chains be deployed to investigate the characteristics of a general language learning algorithm
(Triggering Learning Algorithm)?
This book will appeal to linguists, psychologists, cognitive scientists working in language acquisition. It will also interest those involved in computational modelling in linguistics and behavioural science.
Table of Contents
- Chapter 1: Introduction
- PART I: WORDS
- Chapter 2: Lexicalist Connectionism
- Chapter 3: Are SRNs Sufficient for Modelling Language Acquisition?
- Chapter 4: A Distributed, Yet Symbolic Model for Text-to-Speech Processing
- Chapter 5: "Lazy Learning": A Comparison of Natural and Machine Learning of Word Stress
- PART II: WORD FORMATION
- Chapter 6: Statistical and Connectionist Modelling of the Development of Speech Segmentation
- Chapter 7: Learning Word-to-Meaning Mappings
- Chapter 8: Children's Overregularization and its Implication for Cognition
- Chapter 9: The Performance of a Recurrent Network with Short Term Memory Capacity Learning the German -S Plural
- Chapter 19: A Cross-Linguistic Comparison of Single and Dual-Route Models of Inflectional Morphology
- PART III: WORD ORDER
- Chapter 11: Formal Models for Learning in the Principles and Parameters Framework
- Chapter 12: An Output-as-Input Hypothesis for Language Acquisition: Arguments, Model, Evidence
- Volume
-
: pbk ISBN 9780199256686
Description
This book presents recent advances by leading researchers in computational modelling of language acquisition. Sophisticated theoretical models can now be tested using simulation techniques and large corpora of linguistic data. Renewed interest in learning neural networks and the ability to test new solutions to fundamental problems has fuelled debates in an already very active field. The twenty-four authors in this collection of new work have been drawn from
departments of linguistics, cognitive science, psychology, and computer sciene. The book as a whole shows what light may be thrown on fundamental problems when powerful computational techniques are combined with real data.
A central question addressed in the book concerns the extent to which linguistic structure is readily available in the environment. The authors consider the evidence in relation to word boundaries and phonotactic structure, stress patterns, text-to-speech rules, and the mapping of lexical semantics, one author arguing that a child's own output may serve as a key source of linguistic input. Linguistic structure environment relations are central to the debate on the degree to which language
learning is inductive or deductive; this issue is considered here in studies of the acquisition of pluralization and inflectional morphology.
The book examines the power and utility of different modelling formalisms for different problems and approaches: how far, for example, can connectionist models be used as models for language acquisition? To what degree can lexical items and categories be used in the construction of neural network models or Markov chains be deployed to investigate the characteristics of a general language learning algorithm (Triggering Learning Algorithm)?
This book will appeal to linguists, psychologists, and cognitive scientists working in language acquisition. It will also interest those involved in computational modelling in linguistics and behavioural science.
Table of Contents
- 1. Introduction: The Computational Study of Language Acquisition
- PART I: WORDS
- 2. Lexicalist Connectionism
- 3. Are SRNs Sufficient for Modelling Language Acquisition?
- 4. A Distributed, Yet Symbolic Model for Text-to-Speech Processing
- 5. "Lazy Learning": Natural and Machine Learning of Word Stress
- PART II: WORD FORMATION
- 6. Statistical and Connectionist Modelling of the Development of Speech Segmentation
- 7. Learning Word-to-Meaning Mappings
- 8. Children's Overregularization and its Implication for Cognition
- 9. A Recurrent Network with Short-term Memory Capacity Learning the German -S Plural
- 10. Single- and Dual-Route Models of Inflectional Morphology
- PART III: WORD ORDER
- 11. Formal Models for Learning in the Principles and Parameters Framework
- 12. An Output-as-Input Hypothesis in Language Acquisition
- Notes on Contributors
- Addresses
- Index
by "Nielsen BookData"