The What-And-Where Filter: A Spatial Mapping Neural Network for Object Recognition and Image Understanding

The What-and-Where filter forms part of a neural network architecture for spatial mapping, object recognition, and image understanding. The Where fllter responds to an image figure that has been separated from its background. It generates a spatial map whose cell activations simultaneously represent the position, orientation, ancl size of all tbe figures in a scene (where they are). This spatial map may he used to direct spatially localized attention to these image features. A multiscale array of oriented detectors, followed by competitve and interpolative interactions between position, orientation, and size scales, is used to define the Where filter. This analysis discloses several issues that need to be dealt with by a spatial mapping system that is based upon oriented filters, such as the role of cliff filters with and without normalization, the double peak problem of maximum orientation across size scale, and the different self-similar interpolation properties across orientation than across size scale. Several computationally efficient Where filters are proposed. The Where filter rnay be used for parallel transformation of multiple image figures into invariant representations that are insensitive to the figures' original position, orientation, and size. These invariant figural representations form part of a system devoted to attentive object learning and recognition (what it is). Unlike some alternative models where serial search for a target occurs, a What and Where representation can he used to rapidly search in parallel for a desired target in a scene. Such a representation can also be used to learn multidimensional representations of objects and their spatial relationships for purposes of image understanding. The What-and-Where filter is inspired by neurobiological data showing that a Where processing stream in the cerebral cortex is used for attentive spatial localization and orientation, whereas a What processing stream is used for attentive object learning and recognition.

Fusion ARTMAP: An Adaptive Fuzzy Network for Multi-Channel Classification

Fusion ARTMAP is a self-organizing neural network architecture for multi-channel, or multi-sensor, data fusion. Fusion ARTMAP generalizes the fuzzy ARTMAP architecture in order to adaptively classify multi-channel data. The network has a symmetric organization such that each channel can be dynamically configured to serve as either a data input or a teaching input to the system. An ART module forms a compressed recognition code within each channel. These codes, in turn, beco1ne inputs to a single ART system that organizes the global recognition code. When a predictive error occurs, a process called parallel match tracking simultaneously raises vigilances in multiple ART modules until reset is triggered in one of thmn. Parallel match tracking hereby resets only that portion of the recognition code with the poorest match, or minimum predictive confidence. This internally controlled selective reset process is a type of credit assignment that creates a parsimoniously connected learned network.

ARTMAP: Supervised Real-Time Learning and Classification of Nonstationary Data by a Self-Organizing Neural Network

This article introduces a new neural network architecture, called ARTMAP, that autonomously learns to classify arbitrarily many, arbitrarily ordered vectors into recognition categories based on predictive success. This supervised learning system is built up from a pair of Adaptive Resonance Theory modules (ARTa and ARTb) that are capable of self-organizing stable recognition categories in response to arbitrary sequences of input patterns. During training trials, the ARTa module receives a stream {a^(p)} of input patterns, and ARTb receives a stream {b^(p)} of input patterns, where b^(p) is the correct prediction given a^(p). These ART modules are linked by an associative learning network and an internal controller that ensures autonomous system operation in real time. During test trials, the remaining patterns a^(p) are presented without b^(p), and their predictions at ARTb are compared with b^(p). Tested on a benchmark machine learning database in both on-line and off-line simulations, the ARTMAP system learns orders of magnitude more quickly, efficiently, and accurately than alternative algorithms, and achieves 100% accuracy after training on less than half the input patterns in the database. It achieves these properties by using an internal controller that conjointly maximizes predictive generalization and minimizes predictive error by linking predictive success to category size on a trial-by-trial basis, using only local operations. This computation increases the vigilance parameter ρa of ARTa by the minimal amount needed to correct a predictive error at ARTb· Parameter ρa calibrates the minimum confidence that ARTa must have in a category, or hypothesis, activated by an input a^(p) in order for ARTa to accept that category, rather than search for a better one through an automatically controlled process of hypothesis testing. Parameter ρa is compared with the degree of match between a^(p) and the top-down learned expectation, or prototype, that is read-out subsequent to activation of an ARTa category. Search occurs if the degree of match is less than ρa. ARTMAP is hereby a type of self-organizing expert system that calibrates the selectivity of its hypotheses based upon predictive success. As a result, rare but important events can be quickly and sharply distinguished even if they are similar to frequent events with different consequences. Between input trials ρa relaxes to a baseline vigilance pa When ρa is large, the system runs in a conservative mode, wherein predictions are made only if the system is confident of the outcome. Very few false-alarm errors then occur at any stage of learning, yet the system reaches asymptote with no loss of speed. Because ARTMAP learning is self stabilizing, it can continue learning one or more databases, without degrading its corpus of memories, until its full memory capacity is utilized.

A Spectral Network Model of Pitch Perception

A model of pitch perception, called the Spatial Pitch Network or SPINET model, is developed and analyzed. The model neurally instantiates ideas front the spectral pitch modeling literature and joins them to basic neural network signal processing designs to simulate a broader range of perceptual pitch data than previous spectral models. The components of the model arc interpreted as peripheral mechanical and neural processing stages, which arc capable of being incorporated into a larger network architecture for separating multiple sound sources in the environment. The core of the new model transforms a spectral representation of an acoustic source into a spatial distribution of pitch strengths. The SPINET model uses a weighted "harmonic sieve" whereby the strength of activation of a given pitch depends upon a weighted sum of narrow regions around the harmonics of the nominal pitch value, and higher harmonics contribute less to a pitch than lower ones. Suitably chosen harmonic weighting functions enable computer simulations of pitch perception data involving mistuned components, shifted harmonics, and various types of continuous spectra including rippled noise. It is shown how the weighting functions produce the dominance region, how they lead to octave shifts of pitch in response to ambiguous stimuli, and how they lead to a pitch region in response to the octave-spaced Shepard tone complexes and Deutsch tritones without the use of attentional mechanisms to limit pitch choices. An on-center off-surround network in the model helps to produce noise suppression, partial masking and edge pitch. Finally, it is shown how peripheral filtering and short term energy measurements produce a model pitch estimate that is sensitive to certain component phase relationships.

An efficient neural network approach for nanoscale FinFET modelling and circuit simulation

The present paper demonstrates the suitability of artificial neural network (ANN) for modelling of a FinFET in nano-circuit simulation. The FinFET used in this work is designed using careful engineering of source-drain extension, which simultaneously improves maximum frequency of oscillation f(max) because of lower gate to drain capacitance, and intrinsic gain A(V0) = g(m)/g(ds), due to lower output conductance g(ds). The framework for the ANN-based FinFET model is a common source equivalent circuit, where the dependence of intrinsic capacitances, resistances and dc drain current I-d on drain-source V-ds and gate-source V-gs is derived by a simple two-layered neural network architecture. All extrinsic components of the FinFET model are treated as bias independent. The model was implemented in a circuit simulator and verified by its ability to generate accurate response to excitations not used during training. The model was used to design a low-noise amplifier. At low power (J(ds) similar to 10 mu A/mu m) improvement was observed in both third-order-intercept IIP3 (similar to 10 dBm) and intrinsic gain A(V0) (similar to 20 dB), compared to a comparable bulk MOSFET with similar effective channel length. This is attributed to higher ratio of first-order to third-order derivative of I-d with respect to gate voltage and lower g(ds), in FinFET compared to bulk MOSFET. Copyright (C) 2009 John Wiley & Sons, Ltd.

Network Topology, Higher Orders of Stability and Efficiency

Stable networks of order r where r is a natural number refer to those networks that are immune to coalitional deviation of size r or less. In this paper, we introduce stability of a finite order and examine its relation with efficient networks under anonymous and component additive value functions and the component-wise egalitarian allocation rule. In particular, we examine shapes of networks or network architectures that would resolve the conflict between stability and efficiency in the sense that if stable networks assume those shapes they would be efficient and if efficient networks assume those shapes, they would be stable with minimal further restrictions on value functions.

Recurrent Convolutional Network for Video-based Person Re-Identification

In this paper we propose a novel recurrent neural networkarchitecture for video-based person re-identification.Given the video sequence of a person, features are extracted from each frame using a convolutional neural network that incorporates a recurrent final layer, which allows information to flow between time-steps. The features from all time steps are then combined using temporal pooling to give an overall appearance feature for the complete sequence. The convolutional network, recurrent layer, and temporal pooling layer, are jointly trained to act as a feature extractor for video-based re-identification using a Siamese network architecture.Our approach makes use of colour and optical flow information in order to capture appearance and motion information which is useful for video re-identification. Experiments are conduced on the iLIDS-VID and PRID-2011 datasets to show that this approach outperforms existing methods of video-based re-identification.

https://github.com/niallmcl/Recurrent-Convolutional-Video-ReID
Project Source Code

Phase transitions and nonlinear phenomena in neuronal network models

Communication and cooperation between billions of neurons underlie the power of the brain. How do complex functions of the brain arise from its cellular constituents? How do groups of neurons self-organize into patterns of activity? These are crucial questions in neuroscience. In order to answer them, it is necessary to have solid theoretical understanding of how single neurons communicate at the microscopic level, and how cooperative activity emerges. In this thesis we aim to understand how complex collective phenomena can arise in a simple model of neuronal networks. We use a model with balanced excitation and inhibition and complex network architecture, and we develop analytical and numerical methods for describing its neuronal dynamics. We study how interaction between neurons generates various collective phenomena, such as spontaneous appearance of network oscillations and seizures, and early warnings of these transitions in neuronal networks. Within our model, we show that phase transitions separate various dynamical regimes, and we investigate the corresponding bifurcations and critical phenomena. It permits us to suggest a qualitative explanation of the Berger effect, and to investigate phenomena such as avalanches, band-pass filter, and stochastic resonance. The role of modular structure in the detection of weak signals is also discussed. Moreover, we find nonlinear excitations that can describe paroxysmal spikes observed in electroencephalograms from epileptic brains. It allows us to propose a method to predict epileptic seizures. Memory and learning are key functions of the brain. There are evidences that these processes result from dynamical changes in the structure of the brain. At the microscopic level, synaptic connections are plastic and are modified according to the dynamics of neurons. Thus, we generalize our cortical model to take into account synaptic plasticity and we show that the repertoire of dynamical regimes becomes richer. In particular, we find mixed-mode oscillations and a chaotic regime in neuronal network dynamics.

Exciting dynamical behavior in a network of two coupled rings of Chen oscillators

We study exotic patterns appearing in a network of coupled Chen oscillators. Namely, we consider a network of two rings coupled through a “buffer” cell, with Z3×Z5 symmetry group. Numerical simulations of the network reveal steady states, rotating waves in one ring and quasiperiodic behavior in the other, and chaotic states in the two rings, to name a few. The different patterns seem to arise through a sequence of Hopf bifurcations, period-doubling, and halving-period bifurcations. The network architecture seems to explain certain observed features, such as equilibria and the rotating waves, whereas the properties of the chaotic oscillator may explain others, such as the quasiperiodic and chaotic states. We use XPPAUT and MATLAB to compute numerically the relevant states.

Strange Dynamics in a Fractional Derivative of Complex-Order Network of Chaotic Oscillators

We study the peculiar dynamical features of a fractional derivative of complex-order network. The network is composed of two unidirectional rings of cells, coupled through a "buffer" cell. The network has a Z3 × Z5 cyclic symmetry group. The complex derivative Dα±jβ, with α, β ∈ R+ is a generalization of the concept of integer order derivative, where α = 1, β = 0. Each cell is modeled by the Chen oscillator. Numerical simulations of the coupled cell system associated with the network expose patterns such as equilibria, periodic orbits, relaxation oscillations, quasiperiodic motion, and chaos, in one or in two rings of cells. In addition, fixing β = 0.8, we perceive differences in the qualitative behavior of the system, as the parameter c ∈ [13, 24] of the Chen oscillator and/or the real part of the fractional derivative, α ∈ {0.5, 0.6, 0.7, 0.8, 0.9, 1.0}, are varied. Some patterns produced by the coupled system are constrained by the network architecture, but other features are only understood in the light of the internal dynamics of each cell, in this case, the Chen oscillator. What is more important, architecture and/or internal dynamics?

Interference and communications among active network applications

This paper focuses on active networks applications and in particular on the possible interactions among these applications. Active networking is a very promising research field which has been developed recently, and which poses several interesting challenges to network designers. A number of proposals for e±cient active network architectures are already to be found in the literature. However, how two or more active network applications may interact has not being investigated so far. In this work, we consider a number of applications that have been designed to exploit the main features of active networks and we discuss what are the main benefits that these applications may derive from them. Then, we introduce some forms of interaction including interference and communications among applications, and identify the components of an active network architecture that are needed to support these forms of interaction. We conclude by presenting a brief example of an active network application exploiting the concept of interaction.

A multi-agent architecture for intelligent data analysis

The present study introduces a multi-agent architecture designed for doing automation process of data integration and intelligent data analysis. Different from other approaches the multi-agent architecture was designed using a multi-agent based methodology. Tropos, an agent based methodology was used for design. Based on the proposed architecture, we describe a Web based application where the agents are responsible to analyse petroleum well drilling data to identify possible abnormalities occurrence. The intelligent data analysis methods used was the Neural Network.

Architecture for peer-to-peer databases with routing queries using ant colony algorithm and semantic support

In a peer-to-peer network, the nodes interact with each other by sharing resources, services and information. Many applications have been developed using such networks, being a class of such applications are peer-to-peer databases. The peer-to-peer databases systems allow the sharing of unstructured data, being able to integrate data from several sources, without the need of large investments, because they are used existing repositories. However, the high flexibility and dynamicity of networks the network, as well as the absence of a centralized management of information, becomes complex the process of locating information among various participants in the network. In this context, this paper presents original contributions by a proposed architecture for a routing system that uses the Ant Colony algorithm to optimize the search for desired information supported by ontologies to add semantics to shared data, enabling integration among heterogeneous databases and the while seeking to reduce the message traffic on the network without causing losses in the amount of responses, confirmed by the improve of 22.5% in this amount. © 2011 IEEE.

Development of neurofuzzy architecture for solving the N-Queens problem

Neural networks are dynamic systems consisting of highly interconnected and parallel nonlinear processing elements that are shown to be extremely effective in computation. This paper presents an architecture of recurrent neural networks for solving the N-Queens problem. More specifically, a modified Hopfield network is developed and its internal parameters are explicitly computed using the valid-subspace technique. These parameters guarantee the convergence of the network to the equilibrium points, which represent a solution of the considered problem. The network is shown to be completely stable and globally convergent to the solutions of the N-Queens problem. A fuzzy logic controller is also incorporated in the network to minimize convergence time. Simulation results are presented to validate the proposed approach.

Network convergence and QoS for future multimedia services in the VISION project

The emerging use of real-time 3D-based multimedia applications imposes strict quality of service (QoS) requirements on both access and core networks. These requirements and their impact to provide end-to-end 3D videoconferencing services have been studied within the Spanish-funded VISION project, where different scenarios were implemented showing an agile stereoscopic video call that might be offered to the general public in the near future. In view of the requirements, we designed an integrated access and core converged network architecture which provides the requested QoS to end-to-end IP sessions. Novel functional blocks are proposed to control core optical networks, the functionality of the standard ones is redefined, and the signaling improved to better meet the requirements of future multimedia services. An experimental test-bed to assess the feasibility of the solution was also deployed. In such test-bed, set-up and release of end-to-end sessions meeting specific QoS requirements are shown and the impact of QoS degradation in terms of the user perceived quality degradation is quantified. In addition, scalability results show that the proposed signaling architecture is able to cope with large number of requests introducing almost negligible delay.