Parallel Architectures and Neural Networks

Proceedings of the Fourth Italian Workshop

Fourth Italian Workshop

, Vietri sul Mare, Salerno

, 08 – 10 May 1991

https://doi.org/10.1142/1483 | January 1992

Pages: 396

Edited by:
E R Caianiello (University of Salerno, Italy)

View Full Book

Tools

Purchase Save for later

ISBN: 978-981-4555-78-4 (ebook) USD 38.00 Add to cart

Organized and supported by IIASS and other Italian institutions, this Workshop in Vietri sul Mare has now become a permanent annual event. The aim is to foster closer links among all groups and individuals working in Europe in this inter- and multi-disciplinary area. Topics cover models, architectures and applications.

Contents:

Higher-Order Statistics in a Non-stationary Neural Network (G Basti et al.)
Experimentation on a Learning Algorithm (P Caianiello & V di Leo)
Learning Automata with Sigmoidal Networks (P Frasconi et al.)
Modelling of Associative Memories by Energy Shaping (G Martinelli & R Perfetti)
A Parallel Implementation of Neural Models for Pattern Recognition (R Battiti & A M Colla)
A Connectionist Stored Program Architecture (F Bini Verona et al.)
A Digital Implementation of Self-organizing Feature Maps (M Gioiello et al.)
Dynamic Pruning in Artificial Neural Networks (G Orlandi et al.)
Connectionism for Man-machine Dialogues: An Application to Student Modelling (R Battiti et al.)
Neural Networks Extracting General Features of Protein Secondary Structures (M Compiani et al.)
Isolated Word Recognition by Integration of MLP and LVQ2 Networks (L De Giovanni et al.)
Neural Networks for On-Line Analysis of Oil Well Drilling Parameters (L Malferrari et al.)
and other papers

Readership: Neuroscientists, computer scientists, electrical engineers and mathematicians.

No Access

PARALLEL ARCHITECTURES AND NEURAL NETWORKS: FOURTH ITALIAN WORKSHOP

E. R. CAIANIELLO

Pages:1–383

https://doi.org/10.1142/9789814538480

Preview Abstract

The Table of Contents for the full book PDF is as follows:

Foreword
Scientific Committee; Sponsors
Section 1 MODELS
- Higher-Order Statistics in a Non-stationary Neural Network
- An Algorithm for Learning from Positive Examples Classes of Linearly Separable Boolean Functions
- Generalization Ability of Multilayer Perceptrons Trained in the Chebyshev Norm
- Experimentation on a Learning Algorithm
- A Nondeterministic Method for Multivariate Functions Minimization
- On the Use of Past Information in Reinforcement Learning: Some Experimental Results
- True Steepest Descent Methods in MLP-Networks
- Emerging Complex Behaviours in Ecosystems of Neural Networks
- Learning Automata with Sigmoidal Networks
- Neurophysiological and Neuropsychological Data on Visual Representations of Objects
- Modelling of Associative Memories by Energy Shaping
- A Computational Field Theory of Neural Networks Behaviour
- Neural Network with Complex Dynamics
Section 2 ARCHITECTURES
- A Parallel Implementation of Neural Models for Pattern Recognition
- A Connectionist Stored Program Architecture
- A Hybrid Adaptive-Resonance-Theory/Error-Back-Propagation (ART/EBP) Architecture for Supervised Learning
- Multi Level Dynamic Memorization of Synaptic Weights for Analog CMOS VLSI Neural Circuits
- On Parallelizing Self-organizing Neural Networks
- Artificial Neural Networks: Simulation on PiNa2 System
- An Artificial Neural System Based on Pulse Stream Neural Chips
- A Semicustom Approach for Digital Implementation of Neural Nets
- Comparing the Performance of the Generalized Delta Rule with Singular Value Decomposition Based Global Algorithms for Simple Vision Tasks
- A Digital Implementation of Self-organizing Feature Maps
- Dynamic Pruning in Artificial Neural Networks
- Faults in Multilayer Perceptron
Section 3 APPLICATIONS
- Connectionism for Man-Machine Dialogues: An Application to Student Modelling
- Neural Networks Extracting General Features of Protein Secondary Structures
- Isolated Word Recognition by Integration of MLP and LVQ2 Networks
- Neural Networks for On-line Analysis of Oil Well Drilling Parameters
- A Neural Architecture for the Recognition of Cursive Handwriting
- Linear Neural Networks with Hierarchical Structures for Image Compression
Section 4 SHORT COMMUNICATIONS
- Graphical Simulations for Neural Nets
- A Neural Architecture for the Estimate of 3-D Shape Parameters
- A Fuzzy Trainer for Neural Networks
- Stabilization and Size of Attraction Basins in Symmetric Networks
- Tracking Moving Objects by Unsupervised Learning
- An Introduction to the CC-Net
- An Expert System for Quality Assurance Based on Neural Networks
- A Deterministic Boltzmann Machine Implementing the Mapping between Geometric and Conceptual Knowledge in Visual Perception
- A Case Tool for Neural Networks
- Assessment of the Communication in a Multi-DSP Architecture for a Multilayer Perceptron Simulation
- Nested-Rings Architecture for Feed-forward Networks
- A Multi-net Neural Classifier Tailored by means of Test-Set Characterization
- 40-Hz EEG Attenuation Response in the Recognition of Target Stimuli
- Improving Backpropagation Learning with an Optimum Weight Initialization
- A Multiple Network Architecture for Face Recognition
- Method to Improve the Convergence Capability of the Back-propagation Learning Algorithm Applied to a Multilayer Perceptron NN used for Pattern Recognition
- A New Approach to the Travelling Salesman Problem using Kohonen Maps
- The Implications of Anomalous Transport on the Information Transmission in Neuron-like Structures
- Attractor Reconstruction and Local Forecasting in Chaotic Signals
- A Functional Model of the Striatal Matrix Described as a Cellular Neural Net
- Speaker Independent Vowels Classification using MLP
- Memory Phenomena in Hysteretic Systems
- Neuronic Cellular Automata
- Connectionistic Models of Mental Rotation
- A Kohonen Network for the Recognition of Underwater Structures