Multiplicative Neural Network with Thresholds

Leonid Litinskii, Magomed Malsagov

2013

Abstract

The memory of Hopfield-type neural nets is understood as the ground state of the net – a set of configurations providing a global energy minimum. The use of thresholds allows good control over the ground state. It is possible to build multiplicative networks with the degeneracy of the ground state exceeding considerably the dimensionality of the problem (that is, the net memory can be much greater than the dimensionality of the problem). The paper considers the potentials and limitations of the approach.

References

  1. Hertz, J., Krogh, A., Palmer, R., 1991. Introduction to the Theory of Neural Computation. Massachusetts: Addison-Wesley.
  2. Mertens, S., 2001. A Physicist's Approach to Number Partitioning. Theoret. Comput. Sci. 265, 79-108.
  3. Litinskii, L. B., 1999. High-symmetry Hopfield-type neural networks. Theoretical and Mathematical Physics, 118: 107-127.
  4. Davis, M. W., 2007. The geometry and topology of Coxeter groups. Princeton: Princeton University Press.
  5. Kryzhanovsky, B. V., 2007. Expansion of a matrix in terms of external products of configuration vectors. Optical Memory & Neural Networks (Information Optics), 16: 187-199.
Download


Paper Citation


in Harvard Style

Litinskii L. and Malsagov M. (2013). Multiplicative Neural Network with Thresholds . In Proceedings of the 5th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2013) ISBN 978-989-8565-77-8, pages 523-528. DOI: 10.5220/0004629605230528


in Bibtex Style

@conference{ncta13,
author={Leonid Litinskii and Magomed Malsagov},
title={Multiplicative Neural Network with Thresholds},
booktitle={Proceedings of the 5th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2013)},
year={2013},
pages={523-528},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0004629605230528},
isbn={978-989-8565-77-8},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 5th International Joint Conference on Computational Intelligence - Volume 1: NCTA, (IJCCI 2013)
TI - Multiplicative Neural Network with Thresholds
SN - 978-989-8565-77-8
AU - Litinskii L.
AU - Malsagov M.
PY - 2013
SP - 523
EP - 528
DO - 10.5220/0004629605230528