Search

In this paper, we propose a novel computational architecture of memristor-based echo state network (MESN) with the online least mean square (LMS) algorithm. Newman and Watts small-world network is adopted for the topological structure of MESN network with memristive neural synapses. In the MESN network, the state matrix of the reservoir layer, which is obtained by raising the dimension of input data, is utilized as an input of the LMS algorithm to train the output weight matrix on chip. After certain iterations, the resistance value of memristor is adjusted to a constant. Thus, the final weight output matrix is obtained. To verify the effectiveness of the proposed MESN network, car evaluation and short-term power load forecasting are employed with the effect evaluation of the node number and the connectivity degree of the reservoir layer. The research provides a novel way to design neuromorphic computing systems. [ABSTRACT FROM AUTHOR]

The existing results of passivity for neural networks mainly concentrated on first-order derivative of the states, whereas it is also significant to study the passivity with high-order derivative. In this paper, a class of memristor-based inertial neural networks (MINNs) with external inputs and outputs are concerned. First, by choosing appropriate variable transformation, the original networks are rewritten as first-order differential equations. Then a criterion of passivity for the MINNs is presented by nonsmooth analysis and linear matrix inequality (LMI) techniques, which could also result in NP-hard problem since it needs at least exponential-time to solve the passivity condition. Meanwhile, based on the obtained passivity criterion, asymptotic stability criterion is accordingly derived for MINNs. In order to avoid the NP-hard problem, we employ matrix-analysis-techniques with the property that the difference-matrix between the given matrix and the proposed matrix is seminegative definite. Then the time complexity for solving the proposed passivity criterion is reduced to constant-time with respect to the number of LMIs. Robust passivity for MINNs is also studied for bounded uncertain parameters. The MINN widens the application ranges for designing neural networks. Finally, relevant simulation examples are given to show the effectiveness of the obtained results. [ABSTRACT FROM AUTHOR]

In this paper, coexistence and stability of multiple equilibrium points of fractional-order recurrent neural networks are addressed. Several sufficient conditions are derived for ascertaining the existence of \prod i=1^{n(2Ki+1)} equilibrium points ( Ki\geq 0 ) and the local Mittag–Leffler stability of \prod i=1^{n(Ki+1)} equilibrium points of them by using the geometrical properties of activation functions and algebraic properties of nonsingular M -matrix. In contrast with many existing results, the derived results cover both mono-stability and multistability, and the activation functions herein could be nonmonotonic and nonlinear in any open interval. In addition, three numerical examples are elaborated to substantiate the efficacy and characteristics of the theoretical results. [ABSTRACT FROM AUTHOR]

In this paper, we investigate a general class of fuzzy neurodynamic systems with piecewise constant argument of generalized type. Meanwhile, the time-varying input is under consideration. The pseudo-equilibria of this new type of neurodynamic systems is formulated and studied. Several sufficient conditions are obtained to ensure the existence and uniqueness of solutions. In addition, the global output convergence of such neurodynamic system is examined in detail. Several simulation examples are also given to verify the effectiveness of theoretical property and a potential application in analog associative memory. [ABSTRACT FROM PUBLISHER]

This paper presents new theoretical results on the multistability analysis of a class of recurrent neural networks with nonmonotonic activation functions and mixed time delays. Several sufficient conditions are derived for ascertaining the existence of 3^n equilibrium points and the exponential stability of 2^n equilibrium points via state space partition by using the geometrical properties of activation functions and algebraic properties of nonsingular M -matrix. Compared with existing results, the conditions herein are much more computable with one order less linear matrix inequalities. Furthermore, the attraction basins of these exponentially stable equilibrium points are estimated. It is revealed that the attraction basins of the 2^n equilibrium points can be larger than their originally partitioned subspaces. Three numerical examples are elaborated with typical nonmonotonic activation functions to substantiate the efficacy and characteristics of the theoretical results. [ABSTRACT FROM PUBLISHER]