This paper presents a technique for image recognition, reconstruction, and processing using a novel massively parallel system. This device is a physical implementation of a Boltzmann machine type of neural network based on the use of magnetic thin films and opto-magnetic control. Images or patterns in the form of pixel arrays are imposed on the magnetic film using a laser in an external magnetic field. These images are learned and can be recalled later when a similar image is presented. A stored image is recallable even when a partial, noisy, or corrupted version of that image is imposed on the film. The system can also be used for feature detection and image compression. The operation and construction of the physical system is described, together with a discussion of the physical basis for its operation. The authors have developed Monte Carlo style computer simulations of the system for a variety of platforms, including serial workstations and hypercube configured parallel systems. They describe here some of the factors involved in computer simulations of the system, which can be fast and relatively simple in implementation. Simulation results are presented and, in particular, the behavior of the model under simulated annealing in the light of statistical physics is discussed. The simulation itself can be used as a neural network model capable of the functions ascribed to the physical device.

The main thrust of this paper is to present a new algorithm for the analysis of structure in multivariate data point clusters from the standpoint of intrinsic (topological) dimensionality.

This paper provides guidelines for the discretizing of the space variables in continuous systems governed by the second-order partial differential diffusion equation. Such lumping is necessary in all digital and many analog methods. To analyze the truncation error resulting from lumped analog or digital modelling, the authors demonstrate the applicability of an upper-bound criterion combining amplitude and phase errors in the frequency domain. Dimensionless variables are used to compare lumped networks corresponding to one-dimensional continuous systems. Computer results are presented which relate truncation error to frequency for lumped models containing one to ten sections.