Training a network with ternary weights using the CHIR algorithm

Author

Abramson, S. ; Saad, D. ; Marom, E.

Author_Institution

Fac. of Eng., Tel Aviv Univ., Israel

Volume

4

Issue

6

fYear

1993

fDate

11/1/1993 12:00:00 AM

Firstpage

997

Lastpage

1000

Abstract

A modification of the binary weight CHIR algorithm is presented, whereby a zero state is added to the possible binary weight states. This method allows solutions with reduced connectivity to be obtained, by offering disconnections in addition to the excitatory and inhibitory connections. The algorithm has been examined via extensive computer simulations for the restricted cases of parity, symmetry, and teacher problems, which show convergence rates similar to those presented for the binary CHIR2 algorithm, but with reduced connectivity. Moreover, this method expands the set of problems solvable via the binary weight network configuration with no additional parameter requirements

Keywords

convergence; digital simulation; feedforward neural nets; learning (artificial intelligence); CHIR algorithm; binary CHIR2 algorithm; binary weight network configuration; convergence rates; disconnections; excitatory connections; inhibitory connections; parity; reduced connectivity; symmetry; teacher problems; ternary weights; zero state; Animation; Biology computing; Circuits; Computer networks; Computer vision; Hopfield neural networks; Optimized production technology; Pattern recognition; Physics computing; Visual system;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/72.286901

Filename

286901