Heuristic Self-Organization in Problems of Engineering Cybernetics

The systems, or programs, of heuristic selforganization are defined as those which include the generators of random hypotheses, or combinations, and several layers of threshold self-sampling of useful information. The complexity of combinations increases from layer to layer. A known system, Rosenblatt's perceptron, may be taken as an example. The Group Method of Data Handling (GMDH) based on the principles of heuristic self-organization is developed to solve complex problems with large dimensionality when the data sequence is very short. Two examples are given to illustrate how this method applies to problems of predicting random processes and to identifying characteristics of a multiextremum plant.