edited by James Austin (University of York)

RAM-based networks are a class of methods for building pattern recognition systems. Unlike other neural network methods, they train very rapidly and can be implemented in simple hardware. This important book presents an overview of the subject and the latest work by a number of researchers in the field of RAM-based networks.

• RAM-Based Methods:
  • RAM-Based Neural Networks, a Short History (J Austin)
  • From WISARD to MAGNUS: A Family of Weightless Virtual Neural Machines (I Aleksander)
  • A Comparative Study of GSN^f Learning Methods (A C P L F De Carvalho et al.)
  • The Advanced Uncertain Reasoning Architecture, AURA (J Austin et al.)
• Extensions to N-Tuple Theory:
  • Benchmarking N-Tuple Classifier with StatLog Datasets (M Morciniec & R Rohwer)
  • Comparison of Some Methods for Processing “Grey Level” Data in Weightless Networks (R J Mitchell et al.)
  • A Framework for Reasoning About RAM-Based Neural Networks for Image Analysis Applications (G Howells et al.)
  • Cross-Validation and Information Measures for RAM-Based Neural Networks (T M Jørgensen et al.)
  • A Modular Approach to Storage Capacity (P J L Adeodato & J G Taylor)
  • Good-Turning Estimation for the Frequentist N-Tuple Classifier (M Morciniec & R Rohwer)
  • Partially Pre-Calculated Weights for Backpropagation Training of RAM-Based Sigma–Pi Nets (R Neville)
  • Optimisation of RAM Nets Using Inhibition Between Classes (T M Jørgensen)
  • A New Paradigm for RAM-Based Neural Networks (G Howells et al.)
• Applications of RAM-Based Networks:
  • Content Analysis of Document Images Using the ADAM Associative Memory (S E M O'Keefe & J Austin)
  • Texture Image Classification Using N-Tuple Coding of the Zero-Crossing Sketch (L Hepplewhite & T J Stonham)
  • A Compound Eye for a Simple Robotic Insect (J M Bishop et al.)
  • Extracting Directional Information for the Recognition of Fingerprints by pRAM Networks (T G Clarkson & Y Ding)
  • Detection of Spatial and Temporal Relations in a Two-Dimensional Scene Using a Phased Weightless Neural State Machine (P Ntourntoufis & T J Stonham)
  • Combining Two Boolean Neural Networks for Image Classification (A C P L F De Carvalho et al.)
  • Detecting Danger Labels with RAM-Based Neural Networks (C Linneberg et al.)
  • Fast Simulation of a Binary Neural Network on a Message Passing Parallel Computer (T Macek et al.)
  • C-NNAP: A Dedicated Processor for Binary Neural Networks (J V Kennedy et al.)