The Relevance of the Time Domain to Neural Network Models

2012
ISBN: 978-1-4614-0723-2
Verlag: Springer

Buch, Englisch, 226 Seiten, HC runder Rücken kaschiert, Format (B × H): 160 mm x 241 mm, Gewicht: 530 g

Reihe: Springer Series in Cognitive and Neural Systems

A significant amount of effort in neural modeling is directed towards understanding the representation of external objects in the brain. There is also a rapidly growing interest in modeling the intrinsically-generated activity in the brain, as represented by the default mode network hypothesis, and the emergent behavior that gives rise to critical phenomena such as neural avalanches. Time plays a critical role in these intended modeling domains, from the exquisite discriminations in the mammalian auditory system to the precise timing involved in high-end activities such as competitive sports or professional music performance.

The growth in experimental high-throughput neuroscience techniques has allowed the multi-scale acquisition of neural signals, from individual electrode recordings to whole-brain functional magnetic resonance imaging activity, including the ability to manipulate neural signals with optogenetic approaches. This has created a deluge of experimental data, spanning multiple spatial and temporal scales, and posing the enormous challenge of its interpretation in terms of a predictive theory of brain function. In addition, there has been a massive growth in availability of computational power through parallel computing.

The Relevance of the Time Domain to Neural Network Models aims to develop a unified view of how the time domain can be effectively employed in neural network models. The book proposes that conceptual models of neural interaction are required in order to understand the data being collected. Simultaneously, these proposed models can be used to form hypotheses of neural interaction and system behavior that can be neuroscientifically tested. The book concentrates on a crucial aspect of brain modeling: the nature and functional relevance of temporal interactions in neural systems.

This book will appeal to a wide audience consisting of computer scientists and electrical engineers interested in brain-like computational mechanisms, computer architects exploring the development of high-performance computing systems to support these computations, neuroscientists probing the neural code and signaling mechanisms, mathematicians and physicists interested in modeling complex biological phenomena, and graduate students in all these disciplines who are searching for challenging research questions.

Zielgruppe

Research

Autoren/Hrsg.

Cecchi, Guillermo A.

Rao, A. Ravishankar

Fachgebiete

• Interdisziplinäres Wissenschaften Wissenschaften Interdisziplinär Neurowissenschaften, Kognitionswissenschaft
• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz Neuronale Netzwerke
• Medizin | Veterinärmedizin Medizin | Public Health | Pharmazie | Zahnmedizin Klinische und Innere Medizin Neurologie, Klinische Neurowissenschaft

Inhaltsverzeichnis

Acknowledgements Foreword 1. Introduction 2. Adaptation and contraction theory for the synchronization of complex neural networks 3. Temporal Coding is not only about Cooperation - it is also about Competition 4. Using Non-Oscillatory Dynamics to Disambiguate Simultaneous Patterns 5. Functional constraints on network topology via generalized sparse representations 6. Evolution of Time in Neural Networks: From the Present to the Past, and Forward to the Future 7. Synchronization of Coupled Pulse-Type Hardware Neuron Models for CPG Model 8. A Universal Abstract-Time Platform for Real-Time Neural Networks 9. Solving Complex Control Tasks via Simple Rule(s): Using Chaotic Dynamics in a Recurrent Neural Network Model 10. Time scale analysis of neuronal ensemble data used to feed neural network models 11. Simultaneous EEG-fMRI: Integrating Spatial and Temporal Resolution