A high-storage capacity content-addressable memory and its learning algorithm
- Universite Catholique de Louvain, Laboratoire de microelectronique, 1348 Louvain-la-Neuve (BE)
Hopfield's neural networks show retrieval and speed capabilities that make them good candidates for content-addressable memories (CAM's) in problems such as pattern recognition and optimization. This paper presents a new implementation of a VLSI fully interconnected neural network with only two binary memory points per synapse (the connection weights are restricted to three different values: + 1.0 and -1). The small area of single synaptic cells (about 10/sup 4/ {mu}m/sup 2/) allows the implementation of neural networks with more than 500 neurons. Because of the poor storage capability of Hebb's learning rule, especially in VLSI neural networks where the range of the synapse weights is limited by the number of memory points contained in each connection, a new algorithm is proposed for programming a Hopfield neural network as a high-storage capacity CAM. The results of the VLSI circuit programmed with this new algorithm are promising for pattern recognition applications.
- OSTI ID:
- 5172980
- Journal Information:
- IEEE (Institute of Electrical and Electronics Engineers) Transactions on Circuits and Systems; (USA), Vol. 36:5; ISSN 0098-4094
- Country of Publication:
- United States
- Language:
- English
Similar Records
Modular Spiking Neural Circuits for Mapping Long Short-Term Memory on a Neurosynaptic Processor
Pattern recognition using neural networks. Technical report, August 1, 1994--September 11, 1994
Related Subjects
42 ENGINEERING
INTEGRATED CIRCUITS
PARALLEL PROCESSING
COMPUTER NETWORKS
ALGORITHMS
DESIGN
LEARNING
MEMORY DEVICES
NERVE CELLS
OPTIMIZATION
PATTERN RECOGNITION
PROGRAMMING
SIMULATION
ANIMAL CELLS
ELECTRONIC CIRCUITS
MATHEMATICAL LOGIC
MICROELECTRONIC CIRCUITS
SOMATIC CELLS
990200* - Mathematics & Computers
426000 - Engineering- Components
Electron Devices & Circuits- (1990-)