Year: 1983
Type: report
Abstract: ABSTRACT: In this paper we analyse several approaches to the design of Cooperative Algorithms for solving a general problem: That of computing the values of some property over a spatial domain, when these values are constrained (but not uniquely determined) by some observations, and by some a priori knowledge about the nature of the solution (smoothness, for example). Specifically, we discuss the use of: Variational techniques; stochastic approximation methods for global optimization, and linear threshold net-works. Finally, we present a new approach, based on the interconnection of Winner-take-all networks, for which it is possible to establish precise convergence results, including bounds on the rate of convergence. A.I. Laboratory Working Papers are produced for internal circulation, and may contain information that is, for example, too preliminary or too detailed for formal publication. It is not intended that they should be considered papers to which reference can be made in the literature.
Author: José L. Marroquín
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W1487134871