Simulation of the Neocognitron on a CCD Parallel Processing Architecture
نویسندگان
چکیده
The neocognitron is a neural network for pattern recognition and feature extraction. An analog CCD parallel processing architecture developed at Lincoln Laboratory is particularly well suited to the computational requirements of shared-weight networks such as the neocognitron, and implementation of the neocognitron using the CCD architecture was simulated. A modification to the neocognitron training procedure, which improves network performance under the limited arithmetic precision that would be imposed by the CCD architecture, is presented.
منابع مشابه
Parallel Face Recognition Processing using Neocognitron Neural Network and GPU with CUDA High Performance Architecture
This chapter presents an implementation of the Neocognitron Neural Network, using a high performance computing architecture based on GPU (Graphics Processing Unit). Neocognitron is an artificial neural network, proposed by Fukushima and collaborators, constituted of several hierarchical stages of neuron layers, organized in two-dimensional matrices called cellular plains. For the high performan...
متن کاملA Reconfigurable Computer REOMP
This work describes a proposal of reconfigurable computer, and their appl ication to hardware implementations of neural networks. Although lhe neural network funclions correspond lo the brain functions, our computer is based on the current technology, which is completely di fferent from the internai structure of the brain based on the neuronal cells. The proposed Reconfigurable Orthogonal Multi...
متن کاملReal Time Dynamic Simulation of Power System Using Multiple Microcomputers
Recent developments in the design and manufacture of microcomputers together with improved simulation techniques make it possible to achieve the speed and accuracy required for the dynamic simulation of power systems in real time. This paper presents some experimental results and outlines new ideas on hardware architecture, mathematical algorithms and software development for this purpose. The ...
متن کاملA Vector Orthogoital Multiprocessor NEOMP and its Use in Neural Network Mapping
A vector Orthogonal Multiprocessar architecture NEOMP and its use by a feedforward artificial neural network, Neocognitron, is described. The proposed architecture is composed by severa! vector processing units, and a scalar control processar, which access the memory modules in a orthogonal fashion. The performance analysis of the architecture is realized, identifying the concurrent computation...
متن کاملOptical Implementation of a Neural Network for Pattern Recognition
HIS REPORT DESCRIBES the construction of a dynamic optical hybrid system for implementing multi-layer neural networks. The communication between neurons is performed by amplitude modulating optical signals with dynamic transmission filters realized with a ferroelectric liquid crystal spatial light modulator (FLC-SLM). A large part of the information processing is thus performed in parallel. The...
متن کامل