In this paper two common methods for nonlinear principal component analysis are compared. These two methods are Auto-associative Neural Network (AANN) and Kernel PCA (KPCA). The performance of these methods in sensor data validation are discussed, finally a methodology which takes advantage of both of these methods is presented. The result is a unique approach to nonlinear component mapping of ...

ABSTRACT This paper describes how to improve the robustness to occlusions in face recognition and detection. We propose a neural network architecture which integrates an auto-associative neural network into a simple classifier. The auto-associative network is employed to recall the original face from a partially occluded face image and to detect the occluded regions in the input image. The orig...

Bidirectional Associative Memories (BAMs) are Artificial Neural Networks frequently utilized in cognitive modeling. While bipolar encoding is commonly used BAMs for optimal performance, binary presents interesting properties. As such, this study introduces a novel transmission function and compares its performance to the conventional function. To evaluate, an auto-association learning task nois...

Nowadays, solving optimizations problems is one of the tasks for intelligent computer systems. Currently, there a problem insufficient efficiency methods (for example, high computing time and/or accuracy). The object research process finding shortest path and establishing associative connections between objects. subject objects based on neural networks with memory network reinforcement training...

In this paper, stochastic bidirectional associative memory neural networks with constant or time-varying delays is considered. Based on a Lyapunov-Krasovskii functional and the stochastic stability analysis theory, we derive several sufficient conditions in order to guarantee the global asymptotically stable in the mean square. Our investigation shows that the stochastic bidirectional associati...

This paper presents a probabilistic approach based on collisions to assess the storage capacity of RAM-based neural networks. The analysis at neuron level provides the basis for evaluation of storage capacity in the architectures. The approach is tested in the GNU and pyramid networks. In the GNU as an auto-associative memory, the theoretical results t well with Braga's experimental data and ar...

Recently, some continuous-time recurrent neural networks have been proposed for associative memories based on optimizing linear or quadratic programming problems. In this paper, a simple and efficient neural network with a single recurrent unit is proposed for realizing associative memories. Compared with the existing neural networks for associative memories, the main advantage of the proposed ...

Recent advances in AutoML have led to automated tools that can compete with machine learning experts on supervised learning tasks. However, current AutoML tools do not yet support modern neural networks effectively. In this work, we present a first version of Auto-Net, which provides automatically-tuned feed-forward neural networks without any human intervention. We report results on datasets f...

in recent years, various time series models have been proposed for financial markets forecasting. in each case, the accuracy of time series forecasting models are fundamental to make decision and hence the research for improving the effectiveness of forecasting models have been curried on. many researchers have compared different time series models together in order to determine more efficient ...