This article proposes a modularization scheme for feedforward networks based on controllable internal representations. Control is achieved by replacing hidden units with pretrained modules that constrain internal patterns of activity to desired subsets. In the case of auto-associative feedforward networks these subsets can be seen as module interfaces. If enough a priori knowledge about a syste...

Associative recall and completion of information is one of the astonishing abilities of intelligent living beings. The search for mechanisms which produce this ability of associative memory yielded a class of computational systems composed of many neuron-like, non-linear units. The neural units are connected to artiicial neural networks. The basic principle of associative information recall is ...

abstract nowadays, due to the environmental uncertainty and rapid development of new technologies, economic variables are often predicted by using less data and short-term timeframes. therefore, prediction methods which require fewer amounts of data are needed. auto regressive integrated moving average (arima) model and artificial neural networks (anns) need large amounts of data to achieve acc...

High-level judgement and decision-making tasks display dynamic bidirectional relationships in which salient cues determine how responses are evaluated by decision-makers, and these responses in turn determine the cues that are considered. In this paper, we propose Kosko’s bidirectional associative memory (BAM) network, a minimal two-layer recurrent neural network, as a mathematically tractable ...

Associative memories are conventionally used to represent data with very simple structure: sets of pairs of vectors. This paper describes a method for representing more complex com-positional structure in distributed representations. The method uses circular convolution to associate items, which are represented by vectors. Arbitrary variable bindings, short sequences of various lengths, simple ...

We present a new approach of using Auto-Associative Neural Networks (AANNs) in the conventional GMM speaker verification framework with i-vector feature extraction and PLDA modeling. In this technique, an i-vector feature extractor is trained using adaptation parameters from a mixture of AANNs. In order to model parts of each speaker’s acoustic space, a training objective function based on post...

When sensors malfunction, control systems become unreliable. Even with the most sophisticated instruments and control algorithms, a control decision based on faulty data will likely lead to incorrect control actions. “Sensor Fault Detection” is usually considered as a subset of fault detection. One of the well known approaches in Fault Detection is the model based approach in which a computatio...

This paper presents a novel methodology to detect a set of more suitable attributes that may potentially contribute to emerging faults of a wind turbine. The set of attributes were selected from one-year historical data for analysis. The methodology uses the k-means clustering method to process outlier data and verifies the clustering results by comparing quartiles of boxplots, and applies the ...

Learning to remember long sequences remains a challenging task for recurrent neural networks. Register memory and attention mechanisms were both proposed to resolve the issue with either high computational cost to retain memory differentiability, or by discounting the RNN representation learning towards encoding shorter local contexts than encouraging long sequence encoding. Associative memory,...