作者：
B. Al-kazemiC.K. Mohan2-177 CST
Department of Electrical Engineering and Computer Science Syracuse University Syracuse NY USA
The multi-phase particle swarm optimization algorithm (MPPSO) is a variant of the particle swarm optimization algorithm. It simultaneously evolves multiple groups of particles that change their search criterion when c...
详细信息
The multi-phase particle swarm optimization algorithm (MPPSO) is a variant of the particle swarm optimization algorithm. It simultaneously evolves multiple groups of particles that change their search criterion when changing the phases, and also incorporates hill-climbing. This paper examines the applicability of MPPSO in training feedforward neural network.
In this paper, the applicability of using a multilayer-perceptron (MLP) network in public key cryptography is investigated. A system using the properties of MLP networks is proposed. The security of the proposed syste...
详细信息
In this paper, the applicability of using a multilayer-perceptron (MLP) network in public key cryptography is investigated. A system using the properties of MLP networks is proposed. The security of the proposed system is subsequently examined.
We attempt to explain in more detail the performance of several novel algorithms for nonlinear neural adaptive filtering. Weight trajectories together with the error surface give a clear understandable representation ...
详细信息
We attempt to explain in more detail the performance of several novel algorithms for nonlinear neural adaptive filtering. Weight trajectories together with the error surface give a clear understandable representation of the family of least mean square (LMS) based, nonlinear gradient descent (NGD), search-then-converge (STC) learning algorithms and the real-time recurrent learning (RTRL) algorithm. Performance is measured on prediction of coloured and nonlinear input. The results are an alternative qualitative representation of different qualitative performance measures for the analysed algorithms. Error surfaces and the adjacent instantaneous prediction errors support the analysis.
In this paper, novel robust learning in adaptive processing of data structures for tree representation based image classification is proposed. The idea of this learning scheme is to optimize the free parameters of the...
详细信息
ISBN:
(纸本)076951695X
In this paper, novel robust learning in adaptive processing of data structures for tree representation based image classification is proposed. The idea of this learning scheme is to optimize the free parameters of the node representation in data structures by using the layer-by-layer least squares method. The vanishing gradient information can be recovered to overcome the learning long-term dependency problem for this adaptive processing.
It is possible to derive a simple FPGA architecture from 1-D cellular automata structures in which a 2-D spatial feedforward network is formed. By permitting each site to take on any possible function in its input spa...
详细信息
It is possible to derive a simple FPGA architecture from 1-D cellular automata structures in which a 2-D spatial feedforward network is formed. By permitting each site to take on any possible function in its input space (through LUT substitution) an interesting new Boolean network concept is produced. It can be viewed as an FPGA, and it can be refined in a number of ways to accommodate the addition of configuration circuitry and registration structures. The interesting features of this FPGA include its low descriptive complexity/high regularity, low interconnect demand, interchangeability of logic/routing resources, and defect tolerance. By exploiting a connection between the Vapnik-Chervonenkis dimension of (at least) low-order LUTs and perceptron neural networks, it is relatively straightforward to model these Boolean networks with equivalent artificial neural networks, which can be trained using traditional approaches, such as the backpropagation algorithm. This paper reviews the derivation of this architecture and demonstrates examples of evolved circuit designs.
Splice sites play a very important role for identification of coding regions from DNA sequences of eukaryotic genomes. The paper proposes a novelty machine learning approach to the detection of splice site location in...
详细信息
ISBN:
(纸本)9810475241
Splice sites play a very important role for identification of coding regions from DNA sequences of eukaryotic genomes. The paper proposes a novelty machine learning approach to the detection of splice site location in DNA sequences. The method is based on a hybrid of a Markov model and neural networks where parameters of the Markov model are learned by neural networks. Our proposed model is trained using a backpropagation algorithm. The experiments in the data set of Rogic show that this model performs well that 86% of acceptor sites and 89% of donor sites are correctly found. These results demonstrate the potential use of our approach.
It will be shown the newest results of a hardware realization of a neural net for fast decision making functions in real time. There is a digital micro core with several functions proceeding of the learning and testin...
详细信息
It will be shown the newest results of a hardware realization of a neural net for fast decision making functions in real time. There is a digital micro core with several functions proceeding of the learning and testing of the net, supervising of training process and computation of some calculations in pre-and post-processing. The patterns are automatically presented to the network. The heart of the classifier is a trainable integrated analog neural network structure. Because of its speed the hardware realization is able to solve real time image recognition problems. Simulations with a standard particle image have shown good results not only for the 2D case, but also for the 3D case. Particle positions are extracted from image series first. In a second step, noisy particles were identified from images.
暂无评论