Cargando…
Natural and artificial intelligence : misconceptions about brains and neural networks /
How does the mind work? How is data stored in the brain? How does the mental world connect with the physical world? The hybrid system developed in this book shows a radically new view on the brain. Briefly, in this model memory remains permanent by changing the homeostasis rebuilding the neuronal or...
| Clasificación: | Libro Electrónico |
|---|---|
| Autor principal: | |
| Formato: | Electrónico eBook |
| Idioma: | Inglés |
| Publicado: |
Amsterdam ; New York :
North-Holland,
1992.
|
| Edición: | New, expanded edition. |
| Temas: | |
| Acceso en línea: | Texto completo |
Tabla de Contenidos:
- Front Cover; Natural and Artificial Intelligence: Misconceptions about Brains and Neural Networks; Copyright Page; How to Read the Book; Summary of the Expanded Sections; Summary of the Book (1986); Table of Contents; LIST OF ILLUSTRATIONS; PROLOGUE; 1. Misconceptions about Brain Principles; Are computers brain models?; Should thought depend on an unifying principle?; Should brains be first studied by reductionist methods?; Should neural networks be homogenous?; Are symbolic models a useless return to mental ideas?; Is logic useful in brains?; Can logical reasoning be false?
- Do brains abstract what they process?Must logic be ""neat"" or not exist?; Are artificial intelligence and neural networks incompatible?; Are simple models always better?; Can neural networks be controlled by clocks?; Is mass action functionally different from centralized control?; Are our behaviors perturbed by noise?; 2. Misconceptions about Neural Networks; Do researchers agree on a definition of ""neural networks""?; Are neural networks new systems?; Are neural networks useful only for noisy data?; Can neural networks only compute likelihood?
- Can noise be filtered out in symbolic systems?Can errors be suppressed when learning is irreversible?; Do we actively avoid repeating wrong decisions?; Is the plasticity of neural maps used for learning?; Is learning a transformation of memory?; Should cognitivism be 'neat'?; Are circular NNs similar to attractor NNs?; Are grand-mother neurons impossible?; Can dense patterns be processed in a neural group?; Can memory be made sparse?; Should feature extractors learn?; Why are latches useful?; Are working hypotheses useful?; Is document retrieval a complex algorithm?
- Is distributed more reliable than redundant?Is large capacity opposed to robustness?; Can learning by backpropagation be used in brains?; Should memory be assigned?; Can captor values be used in NNs?; Is topographical organization useful everywhere?; 3. Misconceptions about Brains; Should NNs be called NNNs?; Are all-or-none switches biologically impossible?; Can memory persist in biological organisms?; Can one reliably reason with many false memories?; Can repetition stabilize memory?; Can rehearsal stabilize memory?; Can homeostasis stabilize memory?
- Do neuronal impulses play the role of computer states?Can a weighted sum be computed by a calcium impulse?; Is impulse sparseness not observed in brains?; Do local axons have action potentials?; Can diffusion in the axon of chandelier cells be significant?; Are waking brains desynchronized?; Are oscillations needed for synchronization?; Is the normal behavior of neuronal networks oscillatory?; Is processing in steps seen as oscillations?; Do computer programs have slow rhythms?; Can brain computations be disordered?; Are neurons using only excitation and inhibition?