Foreword |
Neural Pulse Coding |
Spike Timing |
Population Codes |
Hippocampal Place Field |
Hardware Models |
References |
Preface |
The Isaac Newton Institute |
Overview of the Book |
Acknowledgments |
Contributors |
Basic Concepts and Models / Part I: |
Spiking Neurons / 1: |
The Problem of Neural Coding / 1.1: |
Motivation / 1.1.1: |
Rate Codes / 1.1.2: |
Rate as a Spike Count (Average over Time) / 1.1.2.1: |
Rate as a Spike Density (Average over Several Runs) / 1.1.2.2: |
Rate as Population Activity (Average over Several Neurons) / 1.1.2.3: |
Candidate Pulse Codes / 1.1.3: |
Time-to-First-Spike / 1.1.3.1: |
Phase / 1.1.3.2: |
Correlations and Synchrony / 1.1.3.3: |
Stimulus Reconstruction and Reverse Correlation / 1.1.3.4: |
Discussion: Spikes or Rates? / 1.1.4: |
Neuron Models / 1.2: |
Simple Spiking Neuron Model / 1.2.1: |
First Steps towards Coding by Spikes / 1.2.2: |
Threshold-Fire Models / 1.2.3: |
Spike Response Model -- Further Details / 1.2.3.1: |
Integrate-and-Fire Model / 1.2.3.2: |
Models of Noise / 1.2.3.3: |
Conductance-Based Models / 1.2.4: |
Hodgkin-Huxley Model / 1.2.4.1: |
Relation to the Spike Response Model / 1.2.4.2: |
Compartmental Models / 1.2.4.3: |
Rate Models / 1.2.5: |
Conclusions / 1.3: |
Computing with Spiking Neurons / 2: |
Introduction / 2.1: |
A Formal Computational Model for a Network of Spiking Neurons / 2.2: |
McCulloch-Pitts Neurons versus Spiking Neurons / 2.3: |
Computing with Temporal Patterns / 2.4: |
Conincidence Detection / 2.4.1: |
RBF-Units in the Temporal Domain / 2.4.2: |
Computing a Weighted Sum in Temporal Coding / 2.4.3: |
Universal Approximation of Continuous Functions with Spiking Neurons Remarks: / 2.4.4: |
Other Computations with Temporal Patterns in Networks of Spiking Neurons / 2.4.5: |
Computing with a Space-Rate Code / 2.5: |
Computing with Firing Rates / 2.6: |
Computing with Firing Rates and Temporal Correlations / 2.7: |
Networks of Spiking Neurons for Storing and Retrieving Information / 2.8: |
Computing on Spike Trains / 2.9: |
Pulse-Based Computation in VLSI Neural Networks / 2.10: |
Background / 3.1: |
Pulsed Coding: A VLSI Perspective / 3.2: |
Pulse Amplitude Modulation / 3.2.1: |
Pulse Width Modulation / 3.2.2: |
Pulse Frequency Modulation / 3.2.3: |
Phase or Delay Modulation / 3.2.4: |
Noise, Robustness, Accuracy and Speed / 3.2.5: |
A MOSFET Introduction / 3.3: |
Subthreshold Circuits for Neural Networks / 3.3.1: |
Pulse Generation in VLSI / 3.4: |
Pulse Intercommunication / 3.4.1: |
Pulsed Arithmetic in VLSI / 3.5: |
Addition of Pulse Stream Signals / 3.5.1: |
Multiplication of Pulse Stream Signals / 3.5.2: |
MOS Transconductance Multiplier / 3.5.3: |
MOSFET Analog Multiplier / 3.5.4: |
Learning in Pulsed Systems / 3.6: |
Summary and Issues Raised / 3.7: |
Encoding Information in Neuronal Activity / 4: |
Synchronization and Oscillations / 4.1: |
Temporal Binding / 4.3: |
Phase Coding / 4.4: |
Dynamic Range and Firing Rate Codes / 4.5: |
Interspike Interval Variability / 4.6: |
Synapses and Rate Coding / 4.7: |
Summary and Implications / 4.8: |
Implementations / Part II: |
Building Silicon Nervous Systems with Dendritic Tree Neuromorphs / 5: |
Why Spikes? / 5.1: |
Dendritic Processing of Spikes / 5.1.2: |
Tunability / 5.1.3: |
Implementation in VLSI / 5.2: |
Artificial Dendrites / 5.2.1: |
Synapses / 5.2.2: |
Dendritic Non-Linearities / 5.2.3: |
Spike-Generating Soma / 5.2.4: |
Excitability Control / 5.2.5: |
Spike Distribution -- Virtual Wires / 5.2.6: |
Neuromorphs in Action / 5.3: |
Feedback to Threshold-Setting Synapses / 5.3.1: |
Discrimination of Complex Spatio-Temporal Patterns / 5.3.2: |
Processing of Temporally Encoded Information / 5.3.3: |
A Pulse-Coded Communications Infrastructure for Neuromorphic Systems / 5.4: |
Neuromorphic Computational Nodes / 6.1: |
Neuromorphic aVLSI Neurons / 6.3: |
Address Event Representation (AER) / 6.4: |
Implementations of AER / 6.5: |
Silicon Cortex / 6.6: |
Basic Layout / 6.6.1: |
Functional Tests of Silicon Cortex / 6.7: |
An Example Neuronal Network / 6.7.1: |
An Example of Sensory Input to SCX / 6.7.2: |
Future Research on AER Neuromorphic Systems / 6.8: |
Acknowledgements |
Analog VLSI Pulsed Networks for Perceptive Processing / 7: |
Analog Perceptive Nets Communication Requirements / 7.1: |
Coding Information with Pulses / 7.2.1: |
Multiplexing of the Signals Issued by Each Neuron / 7.2.2: |
Non-Arbitered PFM Communication / 7.2.3: |
Analysis of the NAPFM Communication Systems / 7.3: |
Statistical Assumptions / 7.3.1: |
Detection / 7.3.2: |
Detection by Time-Windowing / 7.3.2.1: |
Direct Interpulse Time Measurement / 7.3.2.2: |
Performance / 7.3.3: |
Data Dependency of System Performance / 7.3.3.1: |
Discussion / 7.3.5: |
Detection by Direct Interpulse Time Measurement / 7.3.5.1: |
Address Coding / 7.4: |
Silicon Retina Equipped with the NAPFM Communication System / 7.5: |
Circuit Description / 7.5.1: |
Noise Measurement Results / 7.5.2: |
Projective Field Generation / 7.6: |
Overview / 7.6.1: |
Anisotropic Current Pulse Spreading in a Nonlinear Network / 7.6.2: |
Analysis of the Spatial Response of the Nonlinear Network / 7.6.3: |
Analysis of the Size and Shape of the Bubbles Generable by the Nonlinear Network / 7.6.4: |
Description of the Integrated Circuit for Orientation Enhancement / 7.7: |
System Measurement Results / 7.7.1: |
Other Applications / 7.7.4: |
Weighted Projective Field Generation / 7.7.4.1: |
Complex Projective Field Generation / 7.7.4.2: |
Display Interface / 7.8: |
Conclusion / 7.9: |
Preprocessing for Pulsed Neural VLSI Syste / 8: |
A Sound Segmentation System / 8.1: |
Signal Processing in Analog VLSI / 8.3: |
Continuous Time Active Filters / 8.3.1: |
Sampled Data Active Switched Capacitor (SC) Filters / 8.3.2: |
Sampled Data Active Switched Current (SI) Filters / 8.3.3: |
Palmo -- Pulse Based Signal Processing / 8.3.4: |
Basic Palmo Concepts / 8.4.1: |
The Palmo Signal Representation / 8.4.1.1: |
The Analog Palmo Cell / 8.4.1.2: |
A Palmo Signal Processing System / 8.4.1.3: |
Sources of Harmonic Distortion in a Palmo System / 8.4.1.4: |
A CMOS Analog Palmo Cell Implementation / 8.4.2: |
The Analog Palmo Cell: Details of Circuit Operation / 8.4.2.1: |
Interconnecting Analog Palmo Cells / 8.4.3: |
Results from a Palmo VLSI Device / 8.4.4: |
Digital Processing of Palmo Signals / 8.4.5: |
CMOS Analog Palmo Cell: Performance / 8.4.6: |
Further Work / 8.5: |
Digital Simulation of Spiking Neural Networks / 8.7: |
Implementation Issues of Pulse-Coded Neural Networks / 9.1: |
Discrete-Time Simulation / 9.2.1: |
Requisite Arithmetic Precision / 9.2.2: |
Basic Procedures of Network Computation / 9.2.3: |
Programming Environment / 9.3: |
Concepts of Efficient Simulation / 9.4: |
Mapping Neural Networks on Parallel Computers / 9.5: |
Neuron-Parallelism / 9.5.1: |
Synapse-Parallelism / 9.5.2: |
Pattern-Parallelism / 9.5.3: |
Partitioning of the Network / 9.5.4: |
Performance Study / 9.6: |
Single PE Workstations / 9.6.1: |
Neurocomputer / 9.6.2: |
Parallel Computers / 9.6.3: |
Results of the Performance Study / 9.6.4: |
Design and Analysis of Pulsed Neural Systems / 9.6.5: |
Populations of Spiking Neurons / 10: |
Model / 10.1: |
Population Activity Equation / 10.3: |
Integral Equation for the Dynamics / 10.3.1: |
Normalization / 10.3.2: |
Noise-Free Population Dynamics / 10.4: |
Locking / 10.5: |
Locking Condition / 10.5.1: |
Graphical Interpretation / 10.5.2: |
Transients / 10.6: |
Incoherent Firing / 10.7: |
Determination of the Activity / 10.7.1: |
Stability of Asynchronous Firing / 10.7.2: |
Collective Excitation Phenomena and Their Applications / 10.8: |
Two Variable Formulation of IAF Neurons / 11.1: |
Synchronization of Pulse Coupled Oscillators / 11.2: |
Clustering via Temporal Segmentation / 11.3: |
Limits on Temporal Segmentation / 11.4: |
Image Analysis / 11.5: |
Image Segmentation / 11.5.1: |
Edge Detection / 11.5.2: |
Solitary Waves / 11.6: |
The Importance of Noise / 11.7: |
Acknowledgment / 11.8: |
Computing and Learning with Dynamic Synapses / 12: |
Biological Data on Dynamic Synapses / 12.1: |
Quantitative Models / 12.3: |
On the Computational Role of Dynamic Synapses / 12.4: |
Implications for Learning in Pulsed Neural Nets / 12.5: |
Stochastic Bit-Stream Neural Networks / 12.6: |
Basic Neural Modelling / 13.1: |
Feedforward Networks and Learning / 13.3: |
Probability Level Learning / 13.3.1: |
Bit-Stream Level Learning / 13.3.2: |
Generalization Analysis / 13.4: |
Recurrent Networks / 13.5: |
Applications to Graph Colouring / 13.6: |
Hardware Implementation / 13.7: |
The Stochastic Neuron / 13.7.1: |
Calculating Output Derivatives / 13.7.2: |
Generating Stochastic Bit-Streams / 13.7.3: |
Hebbian Learning of Pulse Timing in the Barn Owl Auditory System / 13.7.4: |
Hebbian Learning / 14.1: |
Review of Standard Formulations / 14.2.1: |
Spike-Based Learning / 14.2.2: |
Example / 14.2.3: |
Learning Window / 14.2.4: |
Barn Owl Auditory System / 14.3: |
The Localization Task / 14.3.1: |
Auditory Localization Pathway / 14.3.2: |
Phase Locking / 14.4: |
Neuron Model / 14.4.1: |
Phase Locking -- Schematic / 14.4.2: |
Simulation Results / 14.4.3: |
Delay Tuning by Hebbian Learning / 14.5: |
Selection of Delays / 14.5.1: |