Some generalizations of the recurrent correlation associative memories

• Grant number: 13/12310-4
• Support Opportunities: Regular Research Grants
• Start date: October 01, 2013
• End date: September 30, 2015
• Field of knowledge: Physical Sciences and Mathematics - Mathematics - Applied Mathematics
• Principal Investigator: Marcos Eduardo Ribeiro Do Valle Mesquita
• Grantee: Marcos Eduardo Ribeiro Do Valle Mesquita
• Host Institution: Instituto de Matemática, Estatística e Computação Científica (IMECC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

Associative memories are mathematical models inspired by the human brain ability to store and recall information by association. Moreover, such as the brain, AMs are also able to recall a certain stored pattern even when the input represents an incomplete or noisy version of a stored item. The class of recurrent correlation associative memories (RCAMs), introduced by Chiueh and Goodman in the early 1990s, includes AM models such as the famous Hopfield network and the exponential correlation associative memory. Many RCAMs exhibit a high-storage capacity and an excellent noise tolerance. Notwithstanding, RCAMs are designed for the storage and recall of bipolar patterns. In this research project, we shall investigate generalizations of RCAMs for real-valued, complex-valued, and multi-valued patterns. The performance of the novel models will be evaluated by means of computational experiments with gray-scale and color images. (AU)

Scientific publications

(5)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

VALLE, MARCOS EDUARDO. Complex-Valued Recurrent Correlation Neural Networks. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, v. 25, n. 9, p. 13-pg., 2014-09-01. (13/12310-4)

VALLE, MARCOS EDUARDO; DE CASTRO, FIDELIS ZANETTI; IEEE. Theoretical and Computational Aspects of Quaternionic Multivalued Hopfield Neural Networks. 2016 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), v. N/A, p. 8-pg., 2016-01-01. (13/12310-4)

VALLE, MARCOS EDUARDO. A Robust Subspace Projection Autoassociative Memory Based on the M-Estimation Method. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, v. 25, n. 7, p. 1372-1377, JUL 2014. (13/12310-4)

VALLE, MARCOS EDUARDO. Complex-Valued Recurrent Correlation Neural Networks. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, v. 25, n. 9, SI, p. 1600-1612, SEP 2014. (13/12310-4)

VALLE, MARCOS EDUARDO; VALENTE, RAUL AMBROZIO. Mathematical Morphology on the Spherical CIELab Quantale with an Application in Color Image Boundary Detection. Journal of Mathematical Imaging and Vision, v. 57, n. 2, p. 183-201, FEB 2017. (13/12310-4)