Bidirectional Associative Memory Networks Research Articles

Visual Mapping for Gender Identification from Facial Images using BAM and DON

Objective: To Develop an intelligent and innovative method to categorize the Gender by focusing facial images. Method: We integrate the characteristics of Bidirectional Associative Memory (BAM) and Deep Octonion Networks (DON) to enhance the Gender detection in real time applications. The developed hybrid model is called Visual Mapping of BAM and DON (VMBAD). To validate the projected system, we make use of 4000 images and 126 different subjects as a data set to train the proposed approach and simultaneously compare our results with the existing methods using the same data set. Findings: The projected technique improves the performance of the system by 3 -5 % in terms of sensitivity, accuracy, and precision when compared with the existing approaches (vide figures 4-7). Novelty: The designed method enhances both the accuracy and precision of image by nearly 4% and 2% respectively when compared with the reported work. Keywords: Artificial Intelligence; Bidirectional Associative Memories; Gender Identification; Deep Octonion Networks; Deep Quaternion Networks

Lyapunov approach to manifolds stability for impulsive Cohen-Grossberg-type conformable neural network models.

In this paper, motivated by the advantages of the generalized conformable derivatives, an impulsive conformable Cohen-Grossberg-type neural network model is introduced. The impulses, which can be also considered as a control strategy, are at fixed instants of time. We define the notion of practical stability with respect to manifolds. A Lyapunov-based analysis is conducted, and new criteria are proposed. The case of bidirectional associative memory (BAM) network model is also investigated. Examples are given to demonstrate the effectiveness of the established results.

Vision-Action Semantic Associative Learning Based on Spiking Neural Networks for Cognitive Robot

Establishing cognitive environments help cognitive robots understand human actions, languages and observed objects. In this paper, a cognitive robotic model based on a novel spiking bidirectional associative memory (BAM) method is presented to establish a cognitive environment. The spiking BAM network uses a supervised spike-based learning rule to learn the relationship between the semantic information of vision and action, where a new coding scheme is proposed to encode and decode vision and action semantic information. Vision semantic information is obtained by the deeply salient shape detection (DSSD) method. Action semantic information is obtained by a spiking neural network (SNN) with the proposed coding method. Experimental results demonstrate that the spiking BAM achieves a good convergence ability and relatively high recall accuracy. Specially, for the cognitive development of the robot, this system can be used to improve the robot’s ability to infer human intentions in its natural interactions with humans.

A reconfigurable bidirectional associative memory network with memristor bridge

Bidirectional associative memories (BAMs) have been extensively applied in auto and heteroassociative learning. However, the research on the hardware implementation of BAM neural networks is relatively few. Memristor, a nanoscale synaptic-like element, provides a new perspective for the circuit implementation of neural networks. In this work, we improve a threshold memristor bridge circuit (T-MBC) and design a novel memristive bidirectional associative memory (MBAM) network circuit on this basis. T-MBC is capable to achieve positive, negative, and zero synaptic weights without any switches, inverters, transistors, and current–voltage conversion processes. The validity of T-MBC is illustrated by PSPICE. MBAM has a simpler structure and gains excellent effects in flexibility, and reconfigurability. Besides, it is shown that MBAM can be trained to recall both disturbed binary images and incomplete/noisy grey-scale images robustly. Meanwhile, a more complex application of emotion classification based on MBAM is also accomplished. Experimental results verify the effectiveness of the MBAM neural network and demonstrate that the MBAM neural network has high recognition accuracy and strong noise immunity.

Bogdanov–Takens and Triple Zero Bifurcations for a Neutral Functional Differential Equations with Multiple Delays

In this paper, a neutral functional differential equation with multiple delays is considered. In a first step, we assumed some sufficient hypotheses to guarantee the existence of the Bogdanov–Takens and the triple-zero bifurcations. In a second step, the normal form of the two bifurcations is obtained by using the reduction on the center manifold and the theory of the normal form. Finally, we applied our study to a class of three-neuron bidirectional associative memory networks, its dynamic behaviors are studied and proved by an example and its numerical simulations.

Design and analysis of associative memories based on external inputs of continuous bidirectional associative networks

This paper presents an extended continuous bidirectional associative memory network (CBAM) and a new ring recurrent network to behave as associative memories with external inputs. The proposed networks are robust in terms of design parameter selection. Some globally exponential stable criteria are derived for the networks with high storage capacity. The approach, by generating networks where the input data are fed via external inputs rather than initial conditions, enables multiple prototype patterns to be retrieved simultaneously. The results improve and extend some previous related works. Recurring to the numerical method, several applicable examples are given to illustrate the effectiveness of the proposed networks.

Oscillatory neural associative memories with synapses based on memristor bridges

An approach to the implementation of electronic associative memories with tunable weights based on the resistor bridges containing memristors--a bidirectional associative memory (BAM) and an associative memory based on the Hopfield network--is proposed. These memories we implement as a networks of coupled phase oscillators. The conditions for the use of the operational amplifier in a comparator mode for implementing the step activation function are determined. It is shown how to use the CMOS transistors switches to control the memristance value. The experiments using LTSPICE models show that for the reference binary images with size 3 × 3 the proposed networks converges to the reference images (and, accordingly, to their inversion) with a random uniform distribution of binary pixel values of the input images. In all experiments we have no error states in spite of the number of reference patterns exceeds the classical estimations for traditional BAM and Hopfield networks.

Handwritten Javanese Character Recognition Using Several Artificial Neural Network Methods

Javanese characters are traditional characters that are used to write the Javanese language. The Javanese language is a language used by many people on the island of Java, Indonesia. The use of Javanese characters is diminishing more and more because of the difficulty of studying the Javanese characters themselves. The Javanese character set consists of basic characters, numbers, complementary characters, and so on. In this research we have developed a system to recognize Javanese characters. Input for the system is a digital image containing several handwritten Javanese characters. Preprocessing and segmentation are performed on the input image to get each character. For each character, feature extraction is done using the ICZ-ZCZ method. The output from feature extraction will become input for an artificial neural network. We used several artificial neural networks, namely a bidirectional associative memory network, a counterpropagation network, an evolutionary network, a backpropagation network, and a backpropagation network combined with chi2. From the experimental results it can be seen that the combination of chi2 and backpropagation achieved better recognition accuracy than the other methods.

Dynamics in Four-Neuron Bidirectional Associative Memory Networks with Inertia and Multiple Delays

Bidirectional associative memory (BAM) networks play an important role in various fields such as optimization, pattern recognition, classification, signal and image processing, parallel computation and associative memory. In this paper, four-neuron BAM networks with inertia and multiple delays are considered. By analyzing the distribution of the eigenvalues of the associated characteristic transcendental equation, local stability criteria are obtained for various system parameters and time delays. By choosing the sum of time delays as a bifurcation parameter, we found that Hopf bifurcation occurs when the sum of time delays passes through a sequence of critical values. Some explicit formulae for determining the stability and the direction of the Hopf bifurcation periodic solutions are obtained by using the normal form theory and center manifold theory. Some numerical simulations are carried out to support theoretical predictions. Our results are new and supplement some previously known studies.

Non-fragile synchronization of memristive BAM networks with random feedback gain fluctuations

This paper investigates the non-fragile synchronization for bidirectional associative memory (BAM) neural networks with time-varying delays. A memristor-based BAM neural network is formulated and sufficient conditions are derived to guarantee its synchronization based on master–slave system approach. The non-fragile observer based feedback controller gains are assumed to have the random fluctuations, two different types of uncertainties which perturb the gains are taken into account. The Lyapunov–Krasovskii stability theory together with linear matrix inequality (LMI) approach is used to derive the delay-dependent criteria to ensure the asymptotic stability of the error system, which guarantees the master system synchronize with the slave system. The expressions for the non-fragile controller can be obtained by solving a set of LMIs using standard MATLAB toolboxes. Finally, numerical example on the chaotic system is presented to illustrate the effectiveness of the theoretical results.

Existence and global exponential stability of anti-periodic solutions of high-order bidirectional associative memory (BAM) networks with time-varying delays on time scales

In this paper, high-order bidirectional associative memory (BAM) networks with time-varying delays on time scales are investigated. With the help of the time scale calculus theory, differential inequality theory, some analysis skills and the Liapunov functional method, a set of sufficient conditions ensuring the existence and exponential stability of anti-periodic solutions for the high-order BAM neural network with time-varying delays are obtained. Our results are new and complement some previously known ones. Moreover, a numerical example is presented to verify the theoretical results.

Dynamical analysis of a delayed six‐neuron BAM network

In this article, a simplified bidirectional associative memory network with delays involving six neurons is considered. By analyzing the distribution of roots of the characteristic equation for linearized system regarding the sum of delay as a bifurcation parameter, we obtain the condition of the occurrence for Hopf bifurcation. It reveals that Hopf bifurcation occurs when the sum of the delay passes through a critical value. The direction and the stability of the bifurcating periodic solution are determined by applying the normal form theory and center manifold theorem. Finally, a numerical example is used to illustrate the validity of theoretical results obtained. © 2015 Wiley Periodicals, Inc. Complexity 21: 9–28, 2016

Novel LMI-Based Condition on Global Asymptotic Stability for a Class of Cohen–Grossberg BAM Networks With Extended Activation Functions

This paper is concerned with global asymptotic stability of a class of Cohen-Grossberg bidirectional associative memory (BAM) neural networks with delays. Under the assumptions that the activation functions only satisfy the so-called extended global Lipschitz condition and the behaved functions only satisfy global Lipschitz condition, we apply linear matrix inequality (LMI) method and homeomorphism theory to propose a new LMI-based sufficient condition for global asymptotic stability of the concerned neural networks. In our results, the extended global Lipschitz condition on the activation functions is less conservative than the assumptions for boundedness and monotonicity and is weaker than the assumption for the general global Lipschitz condition, the global Lipschitz condition on the behaved functions is also less conservative than the assumptions for monotonicity and differentiability in existing papers.

Robust state estimation for uncertain fuzzy bidirectional associative memory networks with time-varying delays

This paper addresses the issue of robust state estimation for a class of fuzzy bidirectional associative memory (BAM) neural networks with time-varying delays and parameter uncertainties. By constructing the Lyapunov–Krasovskii functional, which contains the triple-integral term and using the free-weighting matrix technique, a set of sufficient conditions are derived in terms of linear matrix inequalities (LMIs) to estimate the neuron states through available output measurements such that the dynamics of the estimation error system is robustly asymptotically stable. In particular, we consider a generalized activation function in which the traditional assumptions on the boundedness, monotony and differentiability of the activation functions are removed. More precisely, the design of the state estimator for such BAM neural networks can be obtained by solving some LMIs, which are dependent on the size of the time derivative of the time-varying delays. Finally, a numerical example with simulation result is given to illustrate the obtained theoretical results.

Scansion Text Written in English language and Recognized Printed English Character using Bidirectional Associative Memory Network

Scansion Text Written in English language and Recognized Printed English Character using Bidirectional Associative Memory Network

Application of BAM Network in Fault Diagnosis of Oil-Immerseed Transformer

Bidirectional Associative Memory (BAM) network is presented to analysis the fault of power transformer. In order to improve the classification accuracy, the conception of combination is introduced. The fault diagnosis of power transformer is consisted of 4 BAM networks. The first BAM network is used to classify the normal and fault. The second BAM network is used to classify the heat fault and partial discharge (PD) fault. The third BAM network is used to classify MC-overheating faults in magnetic circuit and EC-overheating faults in electrical circuit. The fourth BAM network is used to classify RSI-discharge faults related to solid insulation, USI-discharge faults unrelated to solid insulation. By comparing with the RBF neural network algorithm for the same 90 input set, we conclude that the BAM network a good classifier for the fault diagnosis of power transformer.

Multiperiodicity of High-Order BAM Networks on Time Scales

This brief paper presents new results about dynamics of high-order bidirectional associative memory (BAM) networks on time scales. Invariant sets on time scales are constructed to orientate regions of orbits. By partitions of index sets, we obtain a unique (periodic) trajectory which is exponentially attractive in each invariant set.

Multiperiodicity Evoked by Periodic External Inputs in Cohen-Grossberg-type BAM Networks with Discrete and Distributed Delays

In this paper, by using contraction mapping theorem, analysis approach and decomposition of solution space, the multiperiodicity issue is discussed for Cohen-Grossberg-type (CG-type) bidirectional associative memory networks (BAMNs) with discrete and distributed delays and a general class of activation functions, where the general class of activation functions consist of nondecreasing functions with saturations including piecewise linear functions with two corner points and standard activation functions as their special cases. It is shown that for any saturation region, if there is a periodic orbit located in it, it must be locally exponentially stable. Then, based on this result, some conditions are derived for ascertaining the (n + m)-neuron CG-type BAMNs can have 2 locally exponentially stable limit cycles located in two saturation regions respectively which are symmetrical. Also, taking account of different saturation regions, results about 2(p + q) (p ≤ m, q ≤ n − 1), 2min{n,m} locally exponentially stable limit cycles can be obtained, where n is the number of the neurons in one layer, m is the number of the neurons in the other layer. Meanwhile, for every locally exponentially stable limit cycle given, the corresponding saturation region can be expressed concretely. Finally, three examples are given to illustrate the effectiveness of the obtained results.

Stability of Bidirectional Associative Memory networks with impulses

Abstract By using Lyapunov functional and some analysis technique, sufficient conditions are obtained for the existence and asymptotic stability of a unique equilibrium of a Bidirectional Associative Memory (BAM) neutral network with Lipschitzian activation functions without assuming their boundedness, monotonicity or differentiability and subjected to impulsive state displacements at fixed instants of time. The sufficient conditions are in terms of the parameters of the network only and are easy to verify.

Robust state estimation for uncertain stochastic bidirectional associative memory networks with time-varying delays

In this paper, we investigate the problem of robust state estimation for uncertain stochastic bidirectional associative memory (BAM) networks with time-varying delays. BAM networks are quite general and include parameter uncertainties, time-varying delays and stochastic disturbances. The activation functions are allowed to satisfy sector-like nonlinearities. The main purpose of investigating this problem is to design a linear estimator to approximate the state of the networks through available measurement outputs. By using the Lyapunov functional method and some stochastic analysis techniques, sufficient conditions are established in terms of linear matrix inequalities (LMIs) which can be checked easily by the Matlab toolbox. It is also shown that the designed state estimator can guarantee the estimation error dynamics to be globally robustly asymptotically stable in the mean square. Two illustrated examples are given to demonstrate the feasibility of the proposed estimation schemes.