Using Artificial Intelligence in Chemistry and Biology: A Practical Guide

by Cartwright; Hugh

• ISBN13: 9780849384127
• ISBN10: 0849384125
• Edition: CD
• Format: Hardcover
• Copyright: 2008-05-05
• Publisher: CRC Press

Summary

Possessing great potential power for gathering and managing data in chemistry, biology, and other sciences, Artificial Intelligence (AI) methods are prompting increased exploration into the most effective areas for their implementation. A comprehensive resource documenting the current state of the science and future directions of the field is required to furnish the working experimental scientist and newcomer alike with the background necessary to utilize these methods.

Table of Contents

Preface	p. xiii
The Author	p. xv
Artificial Intelligence	p. 1
What Is Artificial Intelligence?	p. 2
Why Do We Need Artificial Intelligence and What Can We Do with It?	p. 3
Classification	p. 5
Prediction	p. 5
Correlation	p. 5
Model Creation	p. 6
The Practical Use of Artificial Intelligence Methods	p. 6
Organization of the Text	p. 7
References	p. 8
Artificial Neural Networks	p. 9
Introduction	p. 9
Human Learning	p. 11
Computer Learning	p. 12
The Components of an Artificial Neural Network	p. 14
Nodes	p. 14
Connections	p. 14
Connection Weights	p. 15
Determining the Output from a Node	p. 15
The Input Signal	p. 16
The Activation Function	p. 17
Training	p. 21
More Complex Problems	p. 24
Layered Networks	p. 26
More Flexible Activation Functions	p. 28
Sigmoidal Activation Functions	p. 29
Training a Layered Network: Backpropagation	p. 30
The Mathematical Basis of Backpropagation	p. 32
Learning Rate	p. 35
Momentum	p. 36
Practical Issues	p. 37
Network Geometry and Overfitting	p. 37
The Test Set and Early Stopping	p. 38
Leave One Out	p. 40
Traps and Remedies	p. 40
Network Size	p. 40
Sample Order	p. 40
Ill-Conditioned Networks and the Normalization of Data	p. 41
Random Noise	p. 42
Weight Decay	p. 43
Growing and Pruning Networks	p. 43
Herd Effect	p. 44
Batch or Online Learning	p. 45
Applications	p. 46
Where Do I Go Now?	p. 47
Problems	p. 47
References	p. 48
Self-Organizing Maps	p. 51
Introduction	p. 52
Measuring Similarity	p. 54
Using a Self-Organizing Map	p. 56
Components in a Self-Organizing Map	p. 57
Network Architecture	p. 57
Learning	p. 59
Initialize the Weights	p. 60
Select the Sample	p. 62
Determine Similarity	p. 62
Find the Winning Node	p. 63
Update the Winning Node	p. 64
Update the Neighborhood	p. 65
Repeat	p. 67
Adjustable Parameters in the SOM	p. 71
Geometry	p. 71
Neighborhood	p. 73
Neighborhood Functions	p. 73
Practical Issues	p. 80
Choice of Parameters	p. 80
Visualization	p. 81
Wraparound Maps	p. 85
Maps of Other Shapes	p. 87
Drawbacks of the Self-Organizing Map	p. 88
Applications	p. 89
Where Do I Go Now?	p. 93
Problems	p. 93
References	p. 94
Growing Cell Structures	p. 95
Introduction	p. 96
Growing Cell Structures	p. 98
Training and Evolving a Growing Cell Structure	p. 99
Preliminary Steps	p. 99
Local Success Measures	p. 99
Signal Counter	p. 100
Local Error	p. 100
Best Matching Unit	p. 102
Weights	p. 102
Local Measures Decay	p. 103
Repeat	p. 103
Growing the Network	p. 104
Removing Superfluous Cells	p. 108
Advantages of the Growing Cell Structure	p. 109
Applications	p. 110
Where Do I Go Now?	p. 111
Problems	p. 111
References	p. 111
Evolutionary Algorithms	p. 113
Introduction	p. 113
The Evolution of Solutions	p. 114
Components in a Genetic Algorithm	p. 116
Representation of a Solution in the Genetic Algorithm	p. 117
Operation of the Genetic Algorithm	p. 119
Genetic Algorithm Parameters	p. 120
Initial Population	p. 120
Quality of Solution	p. 121
Selection	p. 124
Crossover	p. 127
Mutation	p. 128
Evolution	p. 130
When Do We Stop?	p. 133
Further Selection and Crossover Strategies	p. 135
Selection Strategies	p. 135
Many-String Tournaments	p. 135
Roulette Wheel Selection	p. 136
Stochastic Remainder Selection	p. 136
Fitness Scaling	p. 138
How Does the Genetic Algorithm Find Good Solutions?	p. 140
Crossover Revisited	p. 142
Uniform Crossover	p. 143
Two-Point Crossover	p. 144
Wraparound Crossover	p. 144
Other Types of Crossover	p. 147
Population Size	p. 148
Focused Mutation	p. 150
Local Search	p. 151
Encoding	p. 151
Real or Binary Coding?	p. 151
Repairing String Damage	p. 155
Fine Tuning	p. 159
Elitism	p. 159
String Fridge	p. 160
Traps	p. 161
Other Evolutionary Algorithms	p. 162
Evolutionary Strategies	p. 162
Genetic Programming	p. 163
Particle Swarm Optimization	p. 166
Applications	p. 168
Where Do I Go Now?	p. 169
Problems	p. 170
References	p. 171
Cellular Automata	p. 173
Introduction	p. 174
Principles of Cellular Automata	p. 175
Components of a Cellular Automata Model	p. 177
The Cells and Their States	p. 177
Transition Rules	p. 179
Neighborhoods	p. 180
The Neighborhood in a One-Dimensional Cellular Automata	p. 180
The Neighborhood in a Two-Dimensional Cellular Automata	p. 181
Theoretical Applications	p. 182
Random Transition Rules	p. 182
Deterministic Transition Rules	p. 185
Voting Rules	p. 185
Practical Considerations	p. 191
Boundaries	p. 191
Additional Parameters	p. 195
Extensions to the Model: Excited States	p. 195
Lattice Gases	p. 197
Applications	p. 19
Where Do I Go Now?	p. 199
Problems	p. 200
References	p. 201
Expert Systems	p. 203
Introduction	p. 204
An Interaction with a Simple Expert System	p. 205
Applicability of Expert Systems	p. 208
Goals of an Expert System	p. 209
The Components of an Expert System	p. 210
The Knowledge Base	p. 211
Factual Knowledge	p. 211
Heuristic Knowledge	p. 213
Rules	p. 214
Case-Based Library	p. 214
Inference Engine and Scheduler	p. 215
The User Interface	p. 215
Knowledge-Base Editor	p. 216
Inference Engine	p. 216
Rule Chaining	p. 218
Forward Chaining	p. 218
Backward Chaining	p. 219
Explanation and Limits to Knowledge	p. 223
Case-Based Reasoning	p. 225
An Expert System for All?	p. 225
Expert System Shells	p. 226
Can I Build an Expert System?	p. 230
Applications	p. 234
Where Do I Go Now?	p. 234
Problems	p. 234
References	p. 235
Fuzzy Logic	p. 237
Introduction	p. 237
Crisp Sets	p. 239
Fuzzy Sets	p. 240
Calculating Membership Values	p. 244
Membership Functions	p. 245
Is Membership the Same as Probability?	p. 248
Hedges	p. 249
How Does a Fuzzy Logic System Work?	p. 250
Input Data	p. 252
Fuzzification of the Input	p. 252
Application of Fuzzy Rules	p. 254
Aggregation	p. 255
Applications	p. 259
Where Do I Go Now?	p. 260
Problems	p. 260
References	p. 261
Learning Classifier Systems	p. 263
Introduction	p. 263
A Basic Classifier System	p. 266
Components of a Classifier System	p. 268
The Environment and Messages from It	p. 269
Input Message List	p. 271
The Classifier List	p. 271
The Output Interface	p. 272
How a Classifier System Works	p. 272
Gather the Input	p. 272
Compare Input Messages with the Classifiers	p. 273
Generate Output Messages	p. 274
Update the Environment	p. 276
Properties of Classifiers	p. 276
Generalist and Specialist Classifiers	p. 276
Classifier Specificity	p. 277
Classifier Strength	p. 278
Learning Classifier System	p. 279
Bids	p. 279
Rewards	p. 281
Learning through Rule Culling	p. 281
Learning through Evolution	p. 283
Coupled Classifiers	p. 284
Taxation	p. 285
Applications	p. 286
Where Do I Go Now?	p. 286
Problems	p. 287
References	p. 287
Evolvable Developmental Systems	p. 289
Introduction and Motivation	p. 289
Relationship between Evolution and Development	p. 290
Production Rules and the Evolution of Digital Circuits	p. 293
Description of Sample Problem	p. 293
Representation of Potential Solution	p. 294
Developmental Procedure	p. 295
Fitness Evaluation	p. 297
Overall Evolution	p. 297
Experimental Results	p. 298
Summary and Challenges	p. 298
Cellular Automata-Like Systems and Evolution of Plane Trusses	p. 299
Developmental Procedure	p. 301
Representation of Potential Solution	p. 302
Overall Evolution	p. 304
Fitness Evaluation	p. 305
Experimental Results	p. 306
Summary and Challenges	p. 306
Cellular Growth and the Evolution of Functions	p. 308
Description of Sample Problem	p. 308
Representation of Potential Solution	p. 309
Developmental Procedure	p. 309
Fitness Evaluation	p. 313
Overall Evolution	p. 312
Experimental Results	p. 312
Summary and Challenges	p. 313
Genomic Regulatory Networks and Modeling Development	p. 314
Description of Sample Problem	p. 314
Representations of Potential Solutions	p. 314
Complex Model of Development	p. 314
Simple Model of Development	p. 315
Developmental Procedures	p. 316
Fitness Evaluation	p. 317
Overall Evolution	p. 319
Extended Example	p. 319
Summary and Challenges	p. 322
Summary	p. 323
Future Challenges and Epilogue	p. 324
Epilogue	p. 326
Acknowledgments	p. 327
References	p. 327
Index	p. 329
Table of Contents provided by Ingram. All Rights Reserved.