Self-organization in Cortical Maps & EM-learning
Francesco Frisone, Pietro G. Morasso, and Luca Perico
University of Genova, Dept. of Informatics, Systems, lecommunications Via Opera Pia 13, 16145 Genova (IT)
Received:April 28, 1998Accepted:August 29, 1998Published:December 20, 1998
Keywords:Self-organization, Hebbian learning, Cortical maps, Population code, EM-learning
Starting from the problem of density estimation, it is shown that Expectation Maximization (EM) learning can be considered a Hebbian mechanism. From this, it is possible to outline a theory of self-organization of cortical maps, which is based on a well-defined optimization process and preserves biologically desirable characteristics such as local computation and uniform treatment of input and lateral connections. A thalamocortical network is described that implements the theory in a fully distributed manner: it uses cortical dynamics for the E-step and Hebbian adaptation of cortico-cortical connections at steady state for the M-step.
Cite this article as:F. Frisone, P. Morasso, and L. Perico, “Self-organization in Cortical Maps & EM-learning,” J. Adv. Comput. Intell. Intell. Inform., Vol.2 No.6, pp. 178-184, 1998.Data files: