Connectionism and the Mind Parallel Processing, Dynamics, and Evolution in Networks

, by Bechtel, William; Abrahamsen, Adele

Note: Supplemental materials are not guaranteed with Rental or Used book purchases.

ISBN: 9780631207122 | 0631207120
Cover: Hardcover
Copyright: 1/21/2002

Rent

(Recommended)

$121.79

Term

Due

Price

Semester

Dec 13

$135.32

Quarter

Sep 5

$128.55

Short Term

Aug 2

$121.79

*This item is part of an exclusive publisher rental program and requires an additional convenience fee. This fee will be reflected in the shopping cart.
Buy New

Currently Available, Usually Ships in 24-48 Hours

$184.87

Connectionism and the Mind provides a clear and balanced introduction to connectionist networks and explores theoretical and philosophical implications. Much of this discussion from the first edition has been updated, and three new chapters have been added on the relation of connectionism to recent work on dynamical systems theory, artificial life, and cognitive neuroscience.Read two of the sample chapters on line:Connectionism and the Dynamical Approach to Cognition: http://www.blackwellpublishing.com/pdf/bechtel.pdfNetworks, Robots, and Artificial Life: http://www.blackwellpublishing.com/pdf/bechtel2.pdf

Preface

xiii

Networks versus Symbol Systems: Two Approaches to Modeling Cognition

(18)

A Revolution in the Making?

(1)

Forerunners of Connectionism: Pandemonium and Perceptrons

(5)

The Allure of Symbol Manipulation

(5)

From logic to artificial intelligence

(3)

From linguistics to information processing

(1)

Using artificial intelligence to simulate human information processing

(1)

The Decline and Re-emergence of Network Models

(3)

Problems with perceptrons

(1)

Re-emergence: The new connectionism

(2)

New Alliances and Unfinished Business

(4)

Notes

(1)

Sources and Suggested Readings

(2)

Connectionist Architectures

(35)

The Flavor of Connectionist Processing: A Simulation of Memory Retrieval

(10)

Components of the model

(2)

Dynamics of the model

(1)

Memory retrieval in the Jets and Sharks network

(1)

The equations

(1)

Illustrations of the dynamics of the model

(1)

Retrieving properties from a name

(2)

Retrieving a name from other properties

(1)

Categorization and prototype formation

(2)

Utilizing regularities

(1)

The Design Features of a Connectionist Architecture

(16)

Patterns of connectivity

(1)

Feedforward networks

(2)

Interactive networks

(1)

Activation rules for units

(1)

Feedforward networks

(2)

Interactive networks: Hopfield networks and Boltzmann machines

(3)

Spreading activation vs. interactive connectionist models

(1)

Learning principles

(2)

Semantic interpretation of connectionist systems

(1)

Localist networks

(1)

Distributed networks

(4)

The Allure of the Connectionist Approach

(6)

Neural plausibility

(1)

Satisfaction of soft constraints

(2)

Graceful degradation

(1)

Content-addressable memory

(2)

Capacity to learn from experience and generalize

(1)

Challenges Facing Connectionist Networks

(1)

Summary

(2)

Notes

(1)

Sources and Recommended Readings

(1)

Learning

(35)

Traditional and Contemporary Approaches to Learning

(3)

Empiricism

(1)

Rationalism

(1)

Contemporary cognitive science

(1)

Connectionist Models of Learning

(25)

Learning procedures for two-layer feedfordward networks

(1)

Training and testing a network

(1)

The Hebbian rule

(2)

the delta rule

(7)

Comparing the Hebbian and delta rules

(1)

Limitations of the delta rule: The XOR problem

(2)

The backpropagation learning procedure for multi-layered networks

(1)

Introducing hidden units and backpropagation learning

(5)

Using backpropagation to solve the XOR problem

(3)

Using backpropagation to train a network to pronounce words

(1)

Some drawbacks of using backpropagation

(1)

Boltzmann learning procedures for non-layered networks

(1)

Competitive learning

(1)

Reinforcement learning

(1)

Some Issues Regarding Learning

(7)

Are connectionist systems associationist?

(2)

Possible roles for innate knowledge

(1)

Networks and the rationalist-empiricist continuum

(1)

Rethinking innateness: Connectionism and emergence

(2)

Notes

(1)

Sources and Suggested Readings

(1)

Pattern Recognition and Cognition

(31)

Networks as Pattern Recognition Devices

(12)

Pattern recognition in two-layer networks

(3)

Pattern recognition in multi-layered networks

(1)

McClelland and Rumelhart's interactive activation model of word recognition

(7)

Evaluating the interactive activation model of word recognition

100

(1)

Generalization and similarity

101

(1)

Extending Pattern Recognition to Higher Cognition

102

(4)

Smolensky's proposal: Reasoning in harmony networks

103

(1)

Margolis's proposal: Cognition as sequential pattern recognition

103

(3)

Logical Inference as Pattern Recognition

106

(11)

What is it to learn logic?

106

(3)

A network for evaluating validity of arguments

109

(3)

Analyzing how a network evaluates arguments

112

(3)

A network for constructing derivations

115

(2)

Beyond Pattern Recognition

117

(3)

Notes

118

(1)

Sources and Suggested Readings

119

(1)

Are Rules Required to Process Representations?

120

(36)

Is Language Use Governed by Rules?

120

(2)

Rumelhart and McClelland's Model of Past-tense Acquisition

122

(13)

A pattern associator with Wickelfeature encodings

122

(4)

Activation function and learning procedure

126

(1)

Overregularization in a simpler network: The rule of 78

127

(3)

Modeling U-shaped learning

130

(3)

Modeling differences between different verb classes

133

(2)

Pinker and Prince's Arguments for Rules

135

(6)

Overview of the critique of Rumelhart and McClelland's model

135

(1)

Putative linguistic inadequacies

136

(3)

Putative behavioral inadequacies

139

(1)

Do the inadequacies reflect inherent limitations of PDP networks?

140

(1)

Accounting for the U-shaped Learning Function

141

(11)

The role of input for children

142

(4)

The role of input for networks: The rule of 78 revisited

146

(2)

Plunkett and Marchman's simulations of past-tense acquisition

148

(4)

Conclusion

152

(4)

Notes

153

(2)

Sources and Suggested Readings

155

(1)

Are Syntactically Structured Representations Needed?

156

(44)

Fodor and Pylyshyn's Critique: The Need for Symbolic Representations with Constituent Structure

156

(7)

The need for compositional syntax and semantics

156

(2)

Connectionist representations lack compositionality

158

(2)

Connectionism as providing mere implementation

160

(3)

First connectionist Response: Explicitly Implementing Rules and Representations

163

(7)

Implementing a Production system in a network

163

(2)

The variable binding problem

165

(1)

Shastri and Ajjanagadde's connectionist model of variable binding

166

(4)

Second Connectionist Response: Implementing Functionally Compositional Representations

170

(8)

Functional vs. concatenative compositionality

170

(1)

Developing compressed representations using Pollack's RAAM networks

171

(4)

Functional compositionality of compressed representations

175

(2)

Performing operations on compressed representations

177

(1)

Third Connectionist Response: Employing Procedural Knowledge with External Symbols

178

(12)

Temporal dependencies in processing language

179

(1)

Achieving short-term memory with simple recurrent networks

180

(1)

Elman's first study: Learning grammatical categories

181

(3)

Elman's second study: Respecting dependency relations

184

(3)

Christiansen's extension: Pushing the limits of SRNs

187

(3)

Using External Symbols to Provide Exact Symbol Processing

190

(4)

Clarifying the Standard: Systematicity and Degree of Generalizability

194

(3)

Conclusion

197

(3)

Notes

198

(1)

Sources and Suggested Readings

199

(1)

Simulating Higher Cognition: A Modular Architecture for Processing Scripts

200

(35)

Overview of Scripts

200

(1)

Overview of Miikkulainen's DISCERN System

201

(2)

Modular Connectionist Architectures

203

(3)

FGREP: An Architecture that Allows the System to Devise Its Own Representations

206

(6)

Why FGREP?

206

(2)

Exploring FGREP in a simple sentence parser

208

(2)

Exploring representations for words in categories

210

(2)

Moving to multiple modules: The DISCERN system

212

(1)

A Self-organizing Lexicon using Kohonen Feature Maps

212

(5)

Innovations in lexical design

212

(1)

Using Kohonen feature maps in DISCERN's lexicon

213

(1)

Orthography: From high-dimensional vector representations to map units

213

(3)

Associative connections: From the orthographic map to the semantic map

216

(1)

Semantics: From map unit to high-dimensional vector representations

216

(1)

Reversing direction: From semantic to orthographic representations

216

(1)

Advantages of Kohonen feature maps

216

(1)

Encoding and Decoding Stories as Scripts

217

(6)

Using recurrent FGREP modules in DISCRN

217

(1)

Using the Sentence Parser and Story Parser to encode stories

218

(3)

Using the Story Generator and Sentence Generator to paraphrase stories

221

(2)

Using the Cue Former and Answer Producer to answer questions

223

(1)

A Connectionist Episodic Memory

223

(5)

Making Kohonen feature maps hierarchical

223

(2)

How role-binding maps become self-organized

225

(1)

How role-binding maps become trace feature maps

225

(3)

Performance: Paraphrasing Stories and Answering Questions

228

(3)

Training and testing DISCERN

228

(1)

Watching DISCERN paraphrase a story

229

(1)

Watching DISCERN answer questions

229

(2)

Evaluating DISCERN

231

(2)

Paths Beyond the First Decade of Connectionism

233

(2)

Notes

234

(1)

Sources and Suggested Readings

234

(1)

Connectionism and the Dynamical Approach to Cognition

235

(47)

Are We on the Road to a Dynamical Revolution?

235

(2)

Basic Concepts of DST: The Geometry of Change

237

(6)

Trajectories in state space: Predators and prey

237

(3)

Bifurcation diagrams and chaos

240

(2)

Embodied networks as coupled dynamical systems

242

(1)

Using Dynamical Systems Tools to Analyze Networks

243

(14)

Discovering limit cycles in network controllers for robotic insects

244

(2)

Discovering multiple attractors in network models of reading

246

(2)

Modeling the semantic pathway

248

(1)

Modeling the phonological pathway

249

(4)

Discovering trajectories in SRNs for sentence processing

253

(3)

Dynamical analyses of learning in networks

256

(1)

Putting Chaos to Work in Networks

257

(7)

Skarda and Freeman's model of the olfactory bulb

257

(3)

Shifting interpretations of ambiguous displays

260

(4)

Is Dynamicism a Competitor to Connectionism?

264

(12)

Van Gelder and Port's critique of classic connectionism

264

(1)

Two styles of modeling

265

(1)

Mechanistic versus covering-law explanations

266

(4)

Representations: Who needs them?

270

(6)

Is Dynamicism Complementary to Connectionism?

276

(4)

Conclusion

280

(2)

Notes

280

(1)

Sources and Suggested Readings

281

(1)

Networks, Robots, and Artificial Life

282

(24)

Robots and the Genetic Algorithm

282

(2)

The robot as an artificial lifeform

282

(1)

The genetic algorithm for simulated evolution

283

(1)

Cellular Automata and the Synthetic Strategy

284

(7)

Langton's vision: The synthetic strategy

284

(2)

Emergent structures from simple beings: Cellular automata

286

(2)

Wolfram's four classes of cellular automata

288

(1)

Langton and 1 at the edge of chaos

289

(2)

Evolution and Learning in Food-seekers

291

(4)

Overview and study 1: Evolution without learning

291

(2)

The Baldwin effect and study 2: Evolution with learning

293

(2)

Evolution and Development in Khepera

295

(5)

Introducing Khepera

295

(1)

The development of phenotypes from genotypes

296

(2)

The evolution of genotypes

298

(1)

Embodied networks: Controlling real robots

298

(2)

The Computational Neuroethology of Robots

300

(1)

When Philosophers Encounter Robots

301

(4)

No Cartesian split in embodied agents?

301

(1)

No representations in subsumption architectures?

302

(1)

No intentionality in robots and Chinese rooms?

303

(1)

No armchair when Dennett does philosophy?

304

(1)

Conclusion

305

(1)

Sources and Suggested Readings

305

(1)

Connectionism and the Brain

306

(43)

Connectionism Meets Cognitive Neuroscience

306

(3)

Four Connectionist Models of Brain Processes

309

(32)

What/Where streams in visual processing

309

(4)

The role of the hippocampus in memory

313

(1)

The basic design and functions of the hippocampal system

313

(2)

Spatial navigation in rats

315

(1)

Spatial versus declarative memory accounts

316

(2)

Declarative memory in humans and monkeys

318

(5)

Simulating dyslexia in network models of reading

323

(1)

Double dissociations in dyslexia

323

(4)

Modeling deep dyslexia

327

(4)

Modeling surface dyslexia

331

(4)

Two pathways versus dual routes

335

(3)

The computational power of modular structure in neocortex

338

(3)

The Neural Implausibility of Many Connectionist Models

341

(5)

Biologically implausible aspects of connectionist networks

342

(1)

How important is neurophysiological plausibility?

343

(3)

Whither Connectionism?

346

(3)

Notes

347

(1)

Sources and Suggested Readings

348

(1)

Appendix A: Notation

349

(1)

Appendix B: Glossary

350

(13)

Bibliography

363

(21)

Name Index

384

(11)

Subject Index

395

Please wait while the item is added to your bag...

FREE SHIPPING

Amazon no longer offers textbook rentals. We do!

Connectionism and the Mind Parallel Processing, Dynamics, and Evolution in Networks

Summary

Author Biography

Table of Contents

Supplemental Materials