A family of neural networks and learning algorithms is introduced: the plastic self-adaptive learning machines (PSALM), together with a new interpretation of these neural networks as hyperpolyhedra in the N-dimensional Euclidean space. These networks self-adapt to a continually changing environment by properly changing the orientation of the faces of a hyperpolyhedron as well as its volume. The current structure of the hyperpolyhedron reflects the structure of the current outside world. The network optimally classifies its noise-distorted excitations into categories, after a competition between all possible categories. New categories can be created, and the old ones can be changed, or be forgotten if they are not used for a long time.>

A modeling methodology that combines ideas from IDEF0 and from data flow and data structured system design is reported. The approach allows the development of hierarchical models for large-scale systems that can be used for documentation as well as for decision-making. The approach is described and applied to the modeling of a manufacturing system. With little modification, the methodology is also recommended for any large system whose structure and operation are not easily determined. Potential applications include biological, physiological, ecological and economic planning and general management systems.>

A class of large-scale decision systems is described, and a multilevel, multi-intelligence information/knowledge/control structure is defined. The description is in a form that features the various levels of signal processing, data fusion, autonomous fuzzy logic and pattern recognition (via neural networks), system knowledge base, and personal expert system, and the way in which these processors are integrated so as to form an operational model for the overall decision system.>

A nonlinear filtering problem is investigated, and the discrete representation theorem is used to synthesize the optimal filter. An optimal filtering algorithm is described, and hybrid techniques are employed to effect several methods of implementing the algorithm on a hybrid computer. These implementations are then compared (via Monte Carlo analysis) on the bases of speed and accuracy. It is concluded that hybrid computation techniques can be used to advantage when studying proposed nonlinear optimal estimation formulations.