Neural Network World; International Journal on Neural and Mass - Parallel Computing and Information Systems

Editorial

The Schizophrenic Brain: A Broken Hermeneutic Circle

A unifying picture to the hermeneutical approach to schizophrenia is given by combining the philosophical and the experimental/computational approaches. Computational models of associative learning and recall in the corticohippocampal system helps to understand the circuits of normal and pathological behavior.

Selective Attention Model of Moving Objects

Tracking moving objects is a vital visual task for the survival of an animal. We describe oscillatory neural network models of visual attention with a central element that can track a moving target among a set of distracters on the screen. At the initial stage, the model forms the focus of attention on an arbitrary object that is considered as a target. Other objects are treated as distracters. We present here two models: 1) synchronisation based model designed as a network of phase oscillato...

Improving Associative Memory in a Network of Spiking Neurons

Associative neural network models are a commonly used methodology when investigating the theory of associative memory in the brain. Comparisons between the mammalian hippocampus and associative memory models of neural networks have been investigated [12]. Biologically based networks are systems built of complex biologically realistic cells with a variety of properties. Here we compare and contrast associative memory function in a network of biologically-based spiking neurons [22] with previou...

Modelling the Stdp Symmetry-to-Asymmetry Transition in the Presence of Gabaergic Inhibition

Experimental studies have shown a symmetry-to-asymmetry transition of the spike-timing dependent plasticity (STDP) curve exists in the proximal stratum radiatum (SR) dendrite of the hippocampal CA1 pyramidal neuron, which is probably due to the presence of GABAergic inhibition [2, 3, 4]. A recent computational model predicted that symmetry-to-asymmetry transition is strongly dependent on the frequency and conductance value of GABA inhibition and that the largest long term potentiation (LTP) v...

From the Inferior Colliculus to a Computational Sound Localization Model

In this paper, we describe a spiking neural network for building an azimuthal sound localization system, which is inspired by the functional organization of the human auditory midbrain up to the inferior colliculus (IC). Our system models two ascending pathways from the cochlear nucleus to the IC: an ITD (Interaural Time Difference) pathway and an ILD (Interaural Level Difference) pathway. We take account of Yin's finding [1] that multiple delay lines only exist in the contralateral medial su...

Intensity Inhomogeneity Correction and Segmentation of Magnetic Resonance Images Using a Multi-Stage Fuzzy Clustering Approach

Intensity inhomogeneity or intensity non-uniformity (INU) is an undesired phenomenon that represents the main obstacle for magnetic resonance (MR) image segmentation and registration methods. Various techniques have been proposed to eliminate or compensate the INU, most of which are embedded into clustering algorithms, and they generally have difficulties when INU reaches high amplitudes. This study proposes a multiple stage fuzzy c-means (FCM) clustering based algorithm for the estimation an...

A Visual Object Recognition System Invariant to Scale and Rotation

We address here the problem of scale and rotation invariant object recognition, making use of a correspondence-based mechanism, in which the identity of an object represented by sensory signals is determined by matching it to a representation stored in memory. The sensory representation is in general affected by various transformations, notably scale and rotation, thus giving rise to the fundamental problem of invariant object recognition. We focus here on a neurally plausible mechanism that ...

A Probabilistic Prediction Method for Object Contour Tracking

We present an approach for probabilistic contour prediction within the framework of an object tracking system. We combine level-set methods for image segmentation with optical flow estimations based on probability distribution functions (pdfs) calculated at each image position. Unlike most recent level-set methods that consider exclusively the sign of the level-set function to determine an object and its background, we introduce a novel interpretation of the value of the level-set function th...

Semantic Adaptation of Neural Network Classifiers in Image Segmentation

Semantic analysis of multimedia content is an ongoing research area that has gained a lot of attention over the last few years. Additionally, machine learning techniques are widely used for multimedia analysis with great success. This work presents a combined approach aiming at the semantic adaptation of neural network classifiers in a multimedia framework. Our proposal is based on a fuzzy reasoning engine which is able to evaluate the outputs and the confidence levels of the neural network c...

Demixing Jazz-Music: Sparse Coding Neural Gas for the Separation of Noisy Overcomplete Sources

We consider the problem of separating noisy overcomplete sources from linear mixtures, i.e., we observe N mixtures of M > N sparse sources. We show that the "Sparse Coding Neural Gas" (SCNG) algorithm [8, 9] can be employed in order to estimate the mixing matrix. Based on the learned mixing matrix the sources are obtained by orthogonal matching pursuit. Using synthetically generated data, we evaluate the influence of (i) the coherence of the mixing matrix, (ii) the noise level, and (iii) t...

Reward-Dependent Sensory Coding in Free-Energy-Based Reinforcement Learning

The free-energy-based reinforcement learning is a new approach to handling high-dimensional states and actions. We investigate its properties using a new experimental platform called the digit floor task. In this task, the high-dimensional pixel data of hand-written digits were directly used as sensory inputs to the reinforcement learning agent. The simulation results showed the robustness of the free-energy-based reinforcement learning method against noise applied in both the training and te...

Clustering Quality and Topology Preservation in Fast Learning Soms

The Self-Organizing Map (SOM) is a popular unsupervised neural network able to provide effective clustering and data visualization for data represented in multidimensional input spaces. In this paper, we describe Fast Learning SOM (FLSOM) which adopts a learning algorithm that improves the performance of the standard SOM with respect to the convergence time in the training phase. We show that FLSOM also improves the quality of the map by providing better clustering quality and topology preser...

Blind Source-Separation in Mixed-Signal Vlsi

This paper describes an implementation of the Kurtosis and InfoMax algorithms for an independent components analysis in mixed-signal CMOS. Our design uses on-chip calibration techniques and local adaptation to compensate for the effect of device mismatch in arithmetic blocks and analog memory cells. We use our design to perform two-input blind source-separation on mixtures of audio signals and mixtures of EEG signals. Our experiments show that the hardware implementation of InfoMax consistent...