Table of Contents

Preface. Contributors. J.L. Johnson, H. Ranganath, G. Kuntimad, and H.J. Caulfield,Pulse-Coupled Neural Networks.H. Li and J. Wang,A Neural Network Model for Optical Flow Computation.F. Unal and N. Tepedelenlioglu,Temporal Pattern Matching Using an Artificial Neural Network.J. Dayhoff, P. Palmadesso, F. Richards, and D.-T. Lin, Patterns of Dynamic Activity and Timing in Neural Network Processing. J. Ghosh, H.-J. Chang, and K. Liano, A Macroscopic Model of Oscillation in Ensembles of Inhibitory and Excitatory Neurons. P. Tito, B. Horne, C.L. Giles, and P. Collingwood, Finite State Machines and Recurrent Neural Networks--Automata and Dynamical Systems Approaches. R. Anderson, Biased Random-Walk Learning: A Neurobiological Correlate to Trial-and-Error. A. Nigrin, Using SONNET 1 to Segment Continuous Sequences of Items. K. Venkatesh, A. Pandya, and S. Hsu,On the Use of High Level Petri Nets in the Modeling of Biological Neural Networks. J. Principe, S. Celebi, B. de Vries, and J. Harris, Locally Recurrent Networks: The Gamma Operator, Properties, and Extensions.Preface. Contributors. J.L. Johnson, H. Ranganath, G. Kuntimad, and H.J. Caulfield, Pulse-Coupled Neural Networks: Introduction. Basic Model. Multiple Pulses. Multiple Receptive Field Inputs. Time Evolution of Two Cells. Space to Time. LinkingWaves and Time Scales. Groups. Invariances. Segmentation. Adaptation. Time to Space. Implementations. Integration into Systems. Concluding Remarks. References. H. Li and J. Wang, A Neural Network Model for Optical Flow Computation: Introduction. Theoretical Background. Discussion on the Reformulation. Choosing Regularization Parameters. A Recurrent Neural Network Model. Experiments. Comparison to Other Work. Summary and Discussion. References. F. Unal and N. Tepedelenlioglu, TemporalPattern Matching Using an Artificial Neural Network: Introduction. Solving Optimization Problems Using the Hopfield Network. Dynamic Time Warping Using Hopfield Network. Computer Simulation Results. Conclusions. References. J. Dayhoff, P. Palmadesso, F. Richards, and D.-T. Lin, Patterns of Dynamic Activity and Timing in Neural Network Processing: Introduction. Dynamic Networks. Chaotic Attractors and Attractor Locking. Developing Multiple Attractors. Attractor Basins and Dynamic Binary Networks. Time Delay Mechanisms and Attractor Training. Timing of Action Potentials in Impulse Trains. Discussion. Acknowledgments. References. J. Ghosh, H.-J. Chang, and K. Liano, A Macroscopic Model of Oscillation in Ensembles of Inhibitory and Excitatory Neurons: Introduction. A Macroscopic Model for Cell Assemblies. Interactions Between Two Neural Groups. Stability of Equilibrium States. Oscillation Frequency Estimation. Experimental Validation. Conclusion. Appendix. References. P. Tito, B. Horne, C.L. Giles, and P. Collingwood, Finite State Machines and Recurrent Neural Networks--Automata and Dynamical Systems Approaches: Introduction. State Machines. Dynamical Systems. Recurrent Neural Network. RNN as a State Machine. RNN as a Collection of Dynamical Systems. RNN with Two State Neurons. Experiments--Learning Loops of FSM. Discussion. References. R. Anderson, Biased Random-Walk Learning: A Neurobiological Correlate to Trial-and-Error: Introduction. Hebb's Rule. Theoretical Learning Rules. Biological Evidence. Conclusions. Acknowledgments. References and Bibliography. A. Nigrin, Using SONNET 1 to Segment Continuous Sequences of Items: Introduction. Learning Isolated and Embedded Spatial Patterns. Storing Items with Decreasing Activity. The LTM Invariance Principle. Using Rehearsal to Process Arbitrarily Long Lists. Implementing the LTM Invariance Principle with an On-Center Off-Surround Circuit. Resetting Items Once They can be Classified. Properties of a Classifying System. Simulations. Discussion. K. Venkatesh, A. Pandya, and S. Hsu, On the Use of High Level Petri Nets in the Modeling of Biological Neural Networks: Introduction. Fundamentals of PNs. Modeling of Biological Neural Systems with High Level PNs. New/Modified Elements Added to HPNs to Model BNNs. Example of a BNN: The Olfactory Bulb. Conclusions. References. J. Principe, S. Celebi, B. de Vries, and J. Harris, Locally Recurrent Networks: The Gamma Operator, Properties, and Extensions: Introduction. Linear Finite Dimensional Memory Structures. The Gamma Neural Network. Applications of the Gamma Memory. Interpretations of the Gamma Memory. Laguerre and Gamma II Memories. Analog VLSI Implementations of the Gamma Filter. Conclusions. References.

Promotional Information

Key Features
* Features neural network architectures on the cutting edge of neural network research
* Brings together highly innovative ideas on dynamical neural networks
* Includes articles written by authors prominent in the neural networks research community
* Provides an authoritative, technically correct presentation of each specific technical area

About the Author

Omid Omidvar is a professor of Computer Science at the University of Computer Science at the University of the District of Columbia, Washington, D.C. He is also a technical director of SPPARC center; a supercomputing facility funded by NSF. He received his Ph.D. from the University of Oklahoma in 1967 and has done extensive work in applications of Neural Networks in Optical Character Recognition and Finger Print for the National Institute of Standards and Technology. Dr. Omidvar has been a consultant to many of the world's most important corporations including IBM, Sun, Gumann, and has completed a five year project for the District of Columbia NASA Consortium in design and performance evaluation of neurocontrollers. Dr. Omidvar is also the Editor-in-Chief of the Journal of Artificial Neural Networks, has been an editor of Progress in Neural Network Series since 1990, and has published a large number of journal and conference publications. In addition to teaching, Dr. Omidvar is also currently working as a computer scientist in the Image Recognition Group, Advanced System Division, at NIST.

Reviews

"Contributors incorporate landmark results on how neural network models have evolved from simple feedforward systems into advanced neural architectures with self-sustained activity patters, simple and complicated oscillations, specialized time elements, and new capabilities for analysis and processing of time-varying signals. Coverage includes the architecture and capabilities of pulse-coupled networks; the relationship between automata and recurrent neural networks; and a putative neurobiological model that correlates with trial-and-error learning." --REFERENCE & RESEARCH BOOK NEWS