Theo Todman's Web Page - Paper Abstracts

Contents

1. Introduction – 1
   • 1.1 Meaning, Understanding, and Thought - 2
   • 1.2 A Road Map - 5
2. The Mind Is a Computer Program - 33
   • 2.1 Evolution as Computation - 47
   • 2.2 The Program of Life - 52
3. The Turing Test, the Chinese Room, and What Computers Can't Do - 67
   • 3.1 The Turing Test - 69
   • 3.2 Semantics vs. Syntax - 75
4. Occam's Razor and Understanding - 79
   • 4.1 Neural Nets and Other Curves - 80
   • 4.2 Minimum Description Length - 93
   • 4.3 Bayesian Statistics - 95
   • 4.4 Summary - 102
5. Optimization - 107
   • 5.1 Hill Climbing - 109
   • 5.2 The Fitness Landscape - 109
   • 5.3 What Good Solutions Look Like - 111
   • 5.4 Back-Propagation - 116
   • 5.5 Why Hill Climbing Works - 118
   • 5.6 Biological Evolution and Genetic Algorithms - 120
   • 5.7 Summary - 125
   • Appendix: Other Potential Problems with the Search for a Turing Machine Input - 126
6. Remarks on Occam's Razor - 129
   • 6.1 Why the Inner Workings of Understanding Are Opaque - 129
   • 6.2 Are Compact Representations Really Necessary? - 135
   • Appendix: The VC Lower Bound - 142
7. Reinforcement Learning - 145
   • 7.1 Reinforcement Learning by Memorizing the State-Space - 146
   • 7.2 Generalization by Building a Compact Evaluation Function - 149
   • 7.3 Why Value Iteration Is Fundamentally Suspect - 153
   • 7.4 Why Neural Nets Are Too Weak a Representation - 155
   • 7.5 Reaction vs. Reflection - 157
   • 7.6 Evolutionary Programming, or Policy Iteration - 159
8. Exploiting Structure - 165
   • 8.1 What Are Objects? - 168
   • 8.2 A Concrete Example: Blocks World - 174
   • 8.3 Games - 187
   • 8.4 Why Hand-Coded Al Programs Are Clueless - 206
   • 8.5 Another Way AI Has Discarded Structure - 208
   • 8.6 Platonism vs. Reality - 211
   • Appendix: Plan Compilation - 212
9. Modules and Metaphors - 215
   • 9.1 Evidence for a Modular Mind - 215
   • 9.2 The Metaphoric Nature of Thought - 220
   • 9.3 The Metaphoric Nature of Thought Reflects Compressed Code - 225
   • 9.4 New Thought and Metaphor on the Fly - 228
   • 9.5 Why a Modular Structure? - 230
10. Evolutionary Programming - 233
    • 10.1 An Economic Model - 240
    • 10.2 The Hayek Machine - 250
    • 10.3 Discussion - 266
11. Intractability - 271
    • 11.1 Hardness - 271
    • 11.2 Polynomial Time Mapping - 281
    • 11.3 So, How Do People Do It? - 286
    • 11.4 Constraint Propagation - 293
12. The Evolution of Learning - 303
    • 12.1 Learning and Development - 308
    • 12.2 Inductive Bias - 316
    • 12.3 Evolution and Inductive Bias - 319
    • 12.4 Evolution's Own Inductive Bias - 320
    • 12.5 The Inductive Bias Evolution Discovers - 323
    • 12.6 The Inductive Bias Built by Evolution into Creatures - 325
    • 12.7 Gene Expression and the Program of Mind - 329
    • 12.8 The Interaction of Learning during Life and Evolution - 331
    • 12.9 Culture: An Even More Powerful Interaction - 335
    • 12.10 A Case Study: Language Learning as an Example of Programmed Inductive Bias - 337
    • 12.11 Grammar Learning and the Baldwin Effect - 343
    • 12.12 Summary - 346
13. Language and the Evolution of Thought - 349
    • 13.1 The Evolution of Behavior from Simple to Complex Creatures - 351
    • 13.2 Review of the Model - 360
    • 13.3 What Is Language? - 365
    • 13.4 Gavagai - 367
    • 13.5 Grammar and Thought - 370
    • 13.6 Nature vs. Nurture: Language and the Divergence between Apes and Modern Humankind - 374
    • 13.7 The Evolution of Language - 378
    • 13.8 Summary - 382
14. The Evolution of Consciousness - 385
    • 14.1 Wanting - 387
    • 14.2 The Self - 403
    • 14.3 Awareness - 408
    • 14.4 Qualia - 424
    • 14.5 Free Will - 426
    • 14.6 Epilogue - 436
15. What Is Thought? – 437