Specification and Implementation of Digital Hopfield-Type Associative Memory with On-Chip Training
نویسندگان
چکیده
This paper addresses the definition of the requirements for the design of a neural network associative memory, with on-chip training, in standard digital CMOS technology. We investigate various learning rules which are integrable in silicon, and we study the associative memory properties of the resulting networks. We also investigate the relationships between the architecture of the circuit and the learning rule, in order to minimize the extra circuitry required for the implementation of training. We describe a 64neuron associative memory with on-chip training, which has been manufactured, and we outline its future extensions. Beyond the application to the specific circuit described in the paper, the general methodology for determining the accuracy requirements can be applied to other circuits and to other auto-associative memory architectures.
منابع مشابه
Specification and implementation of a digital Hopfield-type associative memory with on-chip training
The definition of the requirements for the design of a neural network associative memory, with on-chip training, in standard digital CMOS technology is addressed. Various learning rules that can be integrated in silicon and the associative memory properties of the resulting networks are investigated. The relationships between the architecture of the circuit and the learning rule are studied in ...
متن کاملAnalysing and enhancing the performance of associative memory architectures
This thesis investigates the way in which information about the structure of a set of training data with `natural' characteristics may be used to positively influence the design of associative memory neural network models of the Hopfield type. This is done with a view to reducing the level of connectivity in models of this type. There are three strands to this work. Firstly, an empirical evalua...
متن کاملA comprehensive survey on two associative memory models
This paper surveys two advanced associative memory models[8][5]. The first model was derived from the projection on a closed convex set spanned by patterns. The second model was derived from training weights to improve the error tolerance of the Hopfield network. Both models are designed to resolve the insufficiencies of the Hopfield network. These insufficiencies are loading capacity, limit cy...
متن کاملPulse Density Recurrent Neural Network Systems with Learning Capability Using FPGA
In this paper, we present FPGA recurrent neural network systems with learning capability using the simultaneous perturbation learning rule. In the neural network systems, outputs and internal values are represented by pulse train. That is, analog recurrent neural networks with pulse frequency representation are considered. The pulse density representation and the simultaneous perturbation enabl...
متن کاملAdaptation of Hopfield Associative Memory Parameters in Statistic Training
The paper treats the issue of pattern recognition training in terms of Hopfield associative memory (HAM). The conventional randomization technique is used to determine the exponential extremity of HAM recognition error. The extremity exponent is considered as a function of the training process. In training, the exponent is shown to rise from (1-2p) to (1-2p) where p is the error coefficient at ...
متن کامل