Search

The electric vehicle (EV) and electric vehicle charging station (EVCS) have been widely deployed with the development of large-scale transportation electrifications. However, since charging behaviors of EVs show large uncertainties, the forecasting of EVCS charging power is non-trivial. This paper tackles this issue by proposing a reinforcement learning assisted deep learning framework for the probabilistic EVCS charging power forecasting to capture its uncertainties. Since the EVCS charging power data are not standard time-series data like electricity load, they are first converted to the time-series format. On this basis, one of the most popular deep learning models, the long short-term memory (LSTM) is used and trained to obtain the point forecast of EVCS charging power. To further capture the forecast uncertainty, a Markov decision process (MDP) is employed to model the change of LSTM cell states, which is solved by our proposed adaptive exploration proximal policy optimization (AePPO) algorithm based on reinforcement learning. Finally, experiments are carried out on the real EVCSs charging data from Caltech, and Jet Propulsion Laboratory, USA, respectively. The results and comparative analysis verify the effectiveness and outperformance of our proposed framework., Comment: Accepted by IEEE Transactions on Intelligent Vehicles

This article investigates the stability and synchronization of nonautonomous reaction-diffusion neural networks with general time-varying delays. Compared with the existing works concerning reaction-diffusion neural networks, the main innovation of this article is that the network coefficients are time-varying, and the delays are general (which means that fewer constraints are posed on delays; for example, the commonly used conditions of differentiability and boundedness are no longer needed). By Green's formula and some analytical techniques, some easily checkable criteria on stability and synchronization for the underlying neural networks are established. These obtained results not only improve some existing ones but also contain some novel results that have not yet been reported. The effectiveness and superiorities of the established criteria are verified by three numerical examples.

In this article, we study the consensus issues of multiagent systems (MASs) without any information of the system model by using the reinforcement learning (RL) method and event-based control strategy. First, we design an adaptive event-based consensus control protocol using the local sampled state information so that the consensus errors of all agents are uniformly ultimately bounded. The validity of the above event-triggered adaptive control protocol is confirmed by excluding the Zeno behavior within finite time. Then, based on the RL approach, we present a model-free algorithm to get the feedback gain matrix, and accomplish constructing the adaptive event-triggered control strategy without the knowledge of model information. Distinct with the existing related works, this RL-based event-triggered adaptive control algorithm only relies on the local sampled state information, irrelevant to any model information or global network information. Finally, we provide some examples to demonstrate the validity of the above adaptive event-based consensus algorithm.

The existing results of Lagrange stability for neural networks with distributed time delays are scale-free, which introduces conservativeness naturally. A class of Takagi-Sugeno fuzzy memrisive neural networks (FMNNs) on time scales with discrete time-varying and infinite distributed delays is brought in this paper. First, a new scale-limited Halanay inequality is demonstrated by timescale theory. Next, on the basis of inequality techniques on time scales, some new scale-limited algebraic criteria and linear matrix inequality criteria of Lagrange stability are obtained by comparison strategy and generalized Halanay inequality. All scale-limited sufficient criteria of Lagrange stability for FMNNs not only apply to continuous-time FMNNs and their discrete-time analogues, but also could deal with the arbitrary combination of them. Finally, two numerical simulations are given to verify the validity of the obtained theoretical results.

This article investigates hardware implementation of hierarchical temporal memory (HTM), a brain-inspired machine learning algorithm that mimics the key functions of the neocortex and is applicable to many machine learning tasks. Spatial pooler (SP) is one of the main parts of HTM, designed to learn the spatial information and obtain the sparse distributed representations (SDRs) of input patterns. The other part is temporal memory (TM) which aims to learn the temporal information of inputs. The memristor, which is an appropriate synapse emulator for neuromorphic systems, can be used as the synapse in SP and TM circuits. In this article, a memristor-based SP (MSP) circuit structure is designed to accelerate the execution of the SP algorithm. The presented MSP has properties of modeling both the synaptic permanence and the synaptic connection state within a single synapse, and on-device and parallel learning. Simulation results of statistic metrics and classification tasks on several real-world datasets substantiate the validity of MSP.

The utilization of large-scale distributed renewable energy promotes the development of the multi-microgrid (MMG), which raises the need of developing an effective energy management method to minimize economic costs and keep self energy-sufficiency. The multi-agent deep reinforcement learning (MADRL) has been widely used for the energy management problem because of its real-time scheduling ability. However, its training requires massive energy operation data of microgrids (MGs), while gathering these data from different MGs would threaten their privacy and data security. Therefore, this paper tackles this practical yet challenging issue by proposing a federated multi-agent deep reinforcement learning (F-MADRL) algorithm via the physics-informed reward. In this algorithm, the federated learning (FL) mechanism is introduced to train the F-MADRL algorithm thus ensures the privacy and the security of data. In addition, a decentralized MMG model is built, and the energy of each participated MG is managed by an agent, which aims to minimize economic costs and keep self energy-sufficiency according to the physics-informed reward. At first, MGs individually execute the self-training based on local energy operation data to train their local agent models. Then, these local models are periodically uploaded to a server and their parameters are aggregated to build a global agent, which will be broadcasted to MGs and replace their local agents. In this way, the experience of each MG agent can be shared and the energy operation data is not explicitly transmitted, thus protecting the privacy and ensuring data security. Finally, experiments are conducted on Oak Ridge national laboratory distributed energy control communication lab microgrid (ORNL-MG) test system, and the comparisons are carried out to verify the effectiveness of introducing the FL mechanism and the outperformance of our proposed F-MADRL., Accepted by IEEE Transactions on Neural Networks and Learning Systems

Zero-shot learning casts light on lacking unseen class data by transferring knowledge from seen classes via a joint semantic space. However, the distributions of samples from seen and unseen classes are usually imbalanced. Many zero-shot learning methods fail to obtain satisfactory results in the generalized zero-shot learning task, where seen and unseen classes are all used for the test. Also, irregular structures of some classes may result in inappropriate mapping from visual features space to semantic attribute space. A novel generative mixup networks with semantic graph alignment is proposed in this article to mitigate such problems. To be specific, our model first attempts to synthesize samples conditioned with class-level semantic information as the prototype to recover the class-based feature distribution from the given semantic description. Second, the proposed model explores a mixup mechanism to augment training samples and improve the generalization ability of the model. Third, triplet gradient matching loss is developed to guarantee the class invariance to be more continuous in the latent space, and it can help the discriminator distinguish the real and fake samples. Finally, a similarity graph is constructed from semantic attributes to capture the intrinsic correlations and guides the feature generation process. Extensive experiments conducted on several zero-shot learning benchmarks from different tasks prove that the proposed model can achieve superior performance over the state-of-the-art generalized zero-shot learning.

In this article, master-slave synchronization of reaction-diffusion neural networks (RDNNs) with nondifferentiable delay is investigated via the adaptive control method. First, centralized and decentralized adaptive controllers with state coupling are designed, respectively, and a new analytical method by discussing the size of adaptive gain is proposed to prove the convergence of the adaptively controlled error system with general delay. Then, spatial coupling with adaptive gains depending on the diffusion information of the state is first proposed to achieve the master-slave synchronization of delayed RDNNs, while this coupling structure was regarded as a negative effect in most of the existing works. Finally, numerical examples are given to show the effectiveness of the proposed adaptive controllers. In comparison with the existing adaptive controllers, the proposed adaptive controllers in this article are still effective even if the network parameters are unknown and the delay is nonsmooth, and thus have a wider range of applications.

Inspired by the background adaptive mechanism of eagle vision in different hunting environment, a biological eagle-eye vision adaptation mechanism algorithm is proposed for unmanned aerial vehicle (UAV) to detect the long-distance maritime small target in complex and changeable sea environment. First, the various environment adjustment abilities are summarized according to the physiological structures and characteristics of eagle vision. Second, the mathematical models of glaring adaptation, dim adaptation, color adaptation, and the background adaptation are established based on the background adaptation mechanisms of eagle vision. Last but not least, our proposed biological eagle-eye vision adaptation method and other five comparative experiments are implemented in different scenes, such as dazzling, cloudy, dusk, etc. The results of various evaluation indices show that the information of maritime small target is retained satisfactorily and the background of sea or sky is restrained effectively by the proposed algorithm. The maritime small target detection algorithm can be used for the vision system of UAV operating on the sea. It provides a feasible solution for UAV's remote vision autonomous navigation in changeable sea environment.

This paper investigates a multi-agent system with quantized event-triggered communication mechanisms to reduce the communication expenditure. The event-triggered communication mechanism is proposed to reduce the utilization of communication bandwidth , which lowers the communication expenditure in communication frequency. The quantized communication mechanism quantizes the communication information in the multi-agent system with limited communication capacity. First, a quantized periodic communication mechanism system is proposed, which provides a lower bound of the communication interval for the quantized event-triggered communication system to avoid the Zeno behavior. Then the system with quantized event-triggered communication is proved to be convergent to an optimal solution of distributed constraint optimization . With the quantized communication mechanism, the system can reduce the communication cost in the frequency of communication and the amount of transmission data. The proposed method has a tradeoff between energy saving and precision. Finally, the simulation results with comparisons verify the convergence of the system and exhibit the accuracy with different quantizer densities.

Memristor is considered as a promising synaptic device for neural networks because of its tunable and non-volatile resistance states, which is similar to the biological synapses. In this article, a novel network circuit based on memristor synapses is proposed for bidirectional associative memory with in-situ learning method. An analog neuron circuit is designed to emulate the cubic activation function of neural networks. A memristive synapse circuit is constructed to map both positive and negative weights on a single memristor. Moreover, an in-situ learning circuit fitting memristor's nonlinear characteristic is proposed. Feedback control strategy is incorporated in this learning circuit to adjust the resistance of the memristor and avoid the encoding error of the memristor's write voltage. The performance of the proposed network circuit is verified by the training and recalling simulations. The comparison between the proposed approach and related works is analyzed to demonstrate the advantage of the proposed circuit design.

As we know, the study of complete synchronization of dynamic systems is usually confined to exponential synchronization and power-rate synchronization. Therefore, it is an interesting topic whether there are other complete synchronization methods or give a unified mathematical expression of these complete synchronization types. In this paper, we look into the issue of multi-mode function synchronization (MMFS) for memristive neural networks (MNNs) with two kinds of time-varying delays via event-triggered control. Two types of parameters mismatch in MNNs are considered. One is state-dependent, and by formulating a new Lyapunov functional , we achieve a sufficient criterion for the drive and response MNNs to synchronize in the form of convergence-like function L ( t ) . The other is structure-dependent, which can only realize multi-mode function quasi-synchronization (MMFQS). Matrix measure method and a modified Halanay inequality are used to fulfill the multi-mode function quasi-synchronization between the drive and response MNNs. Conclusively, two numerical examples are simulated to prove the effectiveness of our theoretical results.

This article investigates the global exponential stability and stabilization problems for a class of Takagi–Sugeno (T–S) fuzzy competitive neural networks (NNs). In the considered model, we introduce the T–S fuzzy rule to describe the parametric switching causing by complexity and the vagueness in practical environment. Besides, the effects of reaction diffusions and distributed delays, which inherently exist in circuits of NNs, are also taken into consideration. By using the Lyapunov functional theory and Green formula, several stability criteria in terms of $\mathbb {p}$ -norm are established for the uncompensated fuzzy competitive NNs. Moreover, by designing a fuzzy intermittent controller, the corresponding stabilizability criteria in terms of $\mathbb {p}$ -norm are derived. We also carry out some discussions and comparisons to further show the less conservativeness and wide applicability of the main theorems. Finally, several examples are presented to verify the obtained results.

This article investigates the Lagrange exponential stability and the Lyapunov exponential stability of memristive neural networks with discrete and distributed time-varying delays (DMNNs). By means of inequality techniques, theories of the M-matrix, and the comparison strategy, the Lagrange exponential stability of the underlying DMNNs is considered in the sense of Filippov, and the globally exponentially attractive set is estimated through employing the M-matrix and external input. Especially, when the external input is not concerned, the Lyapunov exponential stability of the corresponding DMNNs is developed immediately in the form of an M-matrix, which contains some published outcomes as special cases. Furthermore, by constructing an M-matrix-based differential system, the Lyapunov exponential stability of the DMNNs is studied, which is less conservative than some existing ones. Finally, three simulation examples are carried out to examine the validness of the theories.

This paper is concerned with the multistability of fractional-order competitive neural networks (FCNNs) with time-varying delays. Based on the division of state space, the equilibrium points (EPs) of FCNNs are given. Several sufficient conditions and criteria are proposed to ascertain the multiple O ( t − α ) -stability of delayed FCNNs. The O ( t − α ) -stability is the extension of Mittag-Leffler stability of fractional-order neural networks, which contains monostability and multistability. Moreover, the attraction basins of the stable EPs of FCNNs are estimated, which shows the attraction basins of the stable EPs can be larger than the divided subsets. These conditions and criteria supplement and improve the previous results. Finally, the results are illustrated by the simulation examples.

As a novel nanoscale device, the memristor has elicited widespread interest in implementing compact and efficient neurocomputing systems in the hardware. In this article, fuzzy deep learning systems using fuzzy memristive modeling methods are presented. One key issue in memristive modeling is the device variation issue due to device-to-device and cycle-to-cycle variations. It is very difficult to distinguish the intermediate memristive states in a multilevel memristor. A fuzzy modeling method is therefore proposed to define the memristive states in a dynamic way. In addition, fuzzy deep learning systems are presented for fuzzy pattern recognition such as image recognition. To the authors’ best knowledge, this is the first work utilizing memristive fuzzy deep learning systems to realize image recognition. The expected input and output in the memristive fuzzy deep learning systems are refined via a bidirectional fuzzy rule. The effectiveness of the proposed fuzzy methods has been verified with comprehensive methods, such as single-layer neural networks, multi-layer neural networks, convolutional neural networks, and $k$ -nearest neighbor method. Another highlight of the proposed fuzzy deep learning system is that there is a great reduction in memory and a significant increase in the speed for image recognition tasks with the same level of testing accuracy.

A differential–algebraic neural network (DANN) with delay (DDANN) is proposed. Firstly, the global existence and uniqueness theorems are established for a DDANN, respectively. Next, a new differential–algebraic inequality is established. Then, a theorem on global exponential stability of DDANN is shown by using this inequality. As an application of DDANN, a very concise criterion on global exponential stability for a neutral-type neural network is given by using DDANNs. Finally, two examples are given to illustrate the theoretical results.

This paper concentrates on the multistability and robustness of complex-valued neural networks (CVNNs) with delays and input perturbation. Firstly, several criteria on the multiple ψ -type stability of CVNNs with time-varying delays are obtained by virtue of ψ -type function and the analytical method. Secondly, several sufficient conditions of robustness have been derived to guarantee the ψ -type stability and boundedness of delayed CVNNs with input perturbation. These obtained results improve and extend the previous results. Finally, one numerical example is provided to show the effectiveness of the theoretical results.

Because it is very expensive to collect a large number of labeled samples to train deep neural networks in certain fields, semi-supervised learning (SSL) researcher has become increasingly important in recent years. There are many consistency regularization-based methods for solving SSL tasks, such as the $$\Pi $$ model and mean teacher. In this paper, we first show through an experiment that the traditional consistency-based methods exist the following two problems: (1) as the size of unlabeled samples increases, the accuracy of these methods increases very slowly, which means they cannot make full use of unlabeled samples. (2) When the number of labeled samples is vary small, the performance of these methods will be very low. Based on these two findings, we propose two methods, metric learning clustering (MLC) and auxiliary fake samples, to alleviate these problems. The proposed methods achieve state-of-the-art results on SSL benchmarks. The error rates are 10.20%, 38.44% and 4.24% for CIFAR-10 with 4000 labels, CIFAR-100 with 10,000 labels and SVHN with 1000 labels by using MLC. For MNIST, the auxiliary fake samples method shows great results in cases with the very few labels.

This article considers both the semi-Markov jumping phenomenon and spatial distribution characteristics when investigating the exponential stabilization of memristive neural networks (MNNs). The introduction of the semi-Markov jumping parameters relaxes the restriction on the sojourn time of Markovian MNNs. To increase the operability while ensuring control effect, a novel event-based spatially pointwise-piecewise switching control scheme is presented under a unified spatial division criterion, in which the pointwise and piecewise control can switch according to the preset event condition for the applicability to different control requirements. Moreover, by constructing a semi-Markov Lyapunov functional and utilizing the properties of the considered cumulative distribution function, the final exponential stabilization criterion and two related corollaries are obtained. Finally, simulation results illustrate the effectiveness and superiority of the proposed control strategy.

Global exponential stability (GES) for impulsive timescale-type nonautonomous neural networks (ITNNNs) with mixed delays is investigated in this article. Discrete time-varying and infinite distributed delays (DTVIDDs) are taken into consideration. First, an improved timescale-type Halanay inequality is proven by timescale theory. Second, several algebraic inequality criteria are demonstrated by constructing impulse-dependent functions and utilizing timescale analytical techniques. Different from the published works, the theoretical results can be applied to GES for ITNNNs and impulsive stabilization design of timescale-type nonautonomous neural networks (TNNNs) with mixed delays. The improved timescale-type Halanay inequality considers time-varying coefficients and DTVIDDs, which improves and extends some existing ones. GES criteria for ITNNNs cover the stability conditions of discrete-time nonautonomous neural networks (NNs) and continuous-time ones, and these theoretical results hold for NNs with discrete-continuous dynamics. The effectiveness of our new theoretical results is verified by two numerical examples in the end.

In this article, we design a new control scheme to investigate the finite-time lag synchronization (FTLS) of inertial neural networks (INNs) with mixed infinite time-varying delays and state-dependent switching. Several novel and easily verified conditions are gained guaranteeing the FTLS of INNs via the finite-time stability theory and nonsmooth analysis. Moreover, it is worth emphasizing that here we directly analyze INNs without using variable substitution, which is different from the reduced-order approach used in correspondingly previous works. At last, a numerical example and an application in image encryption are given to verify the correctness and practicability of the obtained results.

Existing deep neural networks (DNNs) are computationally expensive and memory intensive, which hinder their further deployment in novel nanoscale devices and applications with lower memory resources or strict latency requirements. In this paper, a novel approach to accelerate on-chip learning systems using memristive quantized neural networks (M-QNNs) is presented. A real problem of multilevel memristive synaptic weights due to device-to-device (D2D) and cycle-to-cycle (C2C) variations is considered. Different levels of Gaussian noise are added to the memristive model during each adjustment. Another method of using memristors with binary states to build M-QNNs is presented, which suffers from fewer D2D and C2C variations compared with using multilevel memristors. Furthermore, methods of solving the sneak path issues in the memristive crossbar arrays are proposed. The M-QNN approach is evaluated on two image classification datasets, that is, ten-digit number and handwritten images of mixed National Institute of Standards and Technology (MNIST). In addition, input images with different levels of zero-mean Gaussian noise are tested to verify the robustness of the proposed method. Another highlight of the proposed method is that it can significantly reduce computational time and memory during the process of image recognition.

How to accurately segment a solar photovoltaic panel in an infrared image is an intractable problem due to some unfavorable factors. In this article, an effective approach is proposed for solar photovoltaic panel segmentation from infrared images. In order to alleviate the effect of uneven color distribution, a guided filter-based image-enhancement method is first devised to strengthen the edges of solar photovoltaic panels. Moreover, a two-stage method is proposed to detect the contour lines of solar photovoltaic panels. In our algorithm, first, after a thresholding operation, contours in the images are detected by means of a line-segment detector. Then, a method based on k -means clustering is employed to eliminate lines caused by noise or irrelevant background areas. In addition, a background-subtraction strategy is designed to achieve a more accurate segmentation result by removing the remaining background regions. Experimental results demonstrate the effectiveness and efficiency of the proposed method for the segmentation of solar photovoltaic panels.

This article investigates the leader-follower consensus problem of multiple Euler-Lagrange (EL) systems, where each agent suffers uncertain external disturbances, and the communication links among agents experience faults. Besides, we consider a more general case that only a portion of followers can measure partial components of leader's output and access the dynamic information of leader. The main idea of solving the consensus problem in this article is proceeded in two steps. First, we design an adaptive distributed observer to estimate the full state information of leader in real time with resilience to communication link faults. Second, based on the proposed distributed observer, we propose a proportional-integral (PI) control protocol for each agent to track the trajectory of leader, which is model-independent and robust to uncertain external disturbances. Distinct from the existing leader-follower consensus protocols of multiple EL systems, the proposed distributed observer-based PI consensus protocol in this article is model-independent, which is irrelevant to the structures or features of EL system model. Finally, we present a simulation example to show the resilience of the above adaptive distributed observer and the robustness of the distributed observer-based consensus protocol.

In this article, we investigate the problem of sampled-data robust output feedback control for a class of nonlinear uncertain systems with time-varying disturbance and measurement delay based on continuous-discrete observer. An augmented system that includes the nonlinear uncertain system and disturbance model is first found, and by using the delayed sampled-data output, we then propose a novel predictor-based continuous-discrete observer to estimate the unknown state and disturbance information. After that, in order to attenuate the undesirable influences of nonlinear uncertainties and disturbance, a sampled-data robust output feedback controller is developed based on disturbance/uncertainty estimation and attenuation technique. It shows that under the proposed control method, the states of overall hybrid nonlinear system can converge to a bounded region centered at the origin. The main benefit of the proposed control method is that in the presence of measurement delay, the influences of time-varying disturbance and nonlinear uncertainties can be effectively attenuated with the help of feedback domination method and prediction technique. Finally, the effectiveness of the proposed control method is demonstrated via the simulation results of a numerical example and a practical example.

This article reconsiders synchronization problem of linear complex networks with time-varying delay on time scales. For different types of time scales, aperiodically intermittent control scheme is established by using a matrix-based convex combination method, which has great potential in reducing control consumption and saving communication bandwidth. By employing a common Lyapunov function, aperiodically intermittent controllers are utilized successfully to achieve synchronization of linear delayed complex networks on special time scales onto an isolated node. Next, by constructing a special Lyapunov function with time-varying coefficients, sufficient criteria that consist of two linear matrix inequalities are demonstrated to make linear delayed complex networks on general time scales synchronized onto an isolated system with an exponential convergence rate given in advance. Due to delayed complex networks in this article defined on time scales, the proposed control schemes are applicable to continuous-time networks, their discrete-time forms, and any combination of them. Four numerical examples are offered to highlight the effectiveness and superiority of the proposed aperiodically intermittent control schemes at last.

This article investigates finite-time synchronization of neural networks (NNs) with infinite discrete time-varying delays and discontinuous activations (DDNNs). By virtue of theory of differential inclusions, comparison strategies, and inequality techniques, finite-time synchronization of the underlying DDNNs can be developed via a discontinuous state feedback control law, and the synchronous settling time can be estimated. The delayed state feedback controller and finite-time stability theorem are not employed during the analysis. As a special case, finite-time synchronization of NNs with bounded delays and discontinuous activations is given. Finally, two examples are provided to illustrate the validity of the theories.

This article addresses the multistability and attraction of fractional-order neural networks (FONNs) with unbounded time-varying delays. Several sufficient conditions are given to ensure the coexistence of equilibrium points (EPs) of FONNs with concave-convex activation functions. Moreover, by exploiting the analytical method and the property of the Mittag-Leffler function, it is shown that the multiple Mittag-Leffler stability of delayed FONNs is derived and the obtained criteria do not depend on differentiable time-varying delays. In particular, the criterion of the Mittag-Leffler stability can be simplified to M-matrix. In addition, the estimation of attraction basin of delayed FONNs is studied, which implies that the extension of attraction basin is independent of the magnitude of delays. Finally, three numerical examples are given to show the validity of the theoretical results.

This paper addresses the synchronization of multiple fractional-order recurrent neural networks (RNNs) with time-varying delays under event-triggered communications. Based on the assumption of the existence of strong connectivity or a spanning tree in the communication digraph, two sets of sufficient conditions are derived for achieving event-triggered synchronization. Moreover, an additional condition is derived to preclude Zeno behaviors. As a generalization of existing results, the criteria herein are also applicable to the event-triggered synchronization of multiple integer-order RNNs with or without delays. Two numerical examples are elaborated to illustrate the new results.

This article considers the consensus problem of uncertain multiagent systems, which is addressed by neuroadaptive impulsive control schemes. The proposed control schemes indicate that the communication among agents only occurs impulsively, while the dynamics uncertainty is addressed by adaptive schemes using neural networks. Based on such approaches, two specific control schemes are designed. One is that with impulsive feedback, the control scheme uses continuous-time information, which implies that the adaptive process is continuous over time. Another is that by adopting sampled information, the update of all systems, including the feedbacks on agents, the update of neural networks, and the estimation for uncertainty, can be executed only at impulsive instants. The latter case can reduce the energy cost for communication and control, but extra assistant systems are required. The estimation and consensus prove to be achieved with errors if some conditions are fulfilled. Numerical simulations, including a practical system example, are presented.

Given a novel class instance, the purpose of zero-shot learning (ZSL) is to learn a model to classify the instance by seen samples and semantic information transcending class boundaries. The difficulty lies in how to find a suitable space for zero-shot recognition. The previous approaches use semantic space or visual space as classification space. These methods, which typically learn visual-semantic or semantic-visual mapping and directly exploit the output of the mapping function to measure similarity to classify new categories, do not adequately consider the complementarity and distribution gap of multiple domain information. In this paper, we propose to learn a multidomain information fusion space by a joint learning framework. Specifically, we consider the fusion space as a shared space in which different domain features can be recovered by simple linear transformation. By learning a $n$ -way classifier of fusion space from the seen class samples, we also obtain the discriminative information of the similarity space to make the fusion representation more separable. Extensive experiments on popular benchmark datasets manifest that our approach achieves state-of-the-art performances in both supervised and unsupervised ZSL tasks.

This article is devoted to the cluster synchronization issue of coupled fractional-order neural networks. By introducing the stability theory of fractional-order differential systems and the framework of Filippov regularization, some sufficient conditions are derived for ascertaining the asymptotic and finite-time cluster synchronization of coupled fractional-order neural networks, respectively. In addition, the upper bound of the settling time for finite-time cluster synchronization is estimated. Compared with the existing works, the results herein are applicable for fractional-order systems, which could be regarded as an extension of integer-order ones. A numerical example with different cases is presented to illustrate the validity of theoretical results.

This paper addresses global Mittag-Leffler lag quasi-synchronization (GMLQS) of fractional-order memristor-based neural networks (FMNNs). Firstly, for a general class of fractional-order nonlinear systems with interval uncertainties, a kind of linear feedback pinning controller, which works by feeding back partial state errors to the partial controlled variables, is designed to achieve global Mittag-Leffler ultimate boundedness (GMUB). Correspondingly, a GMUB criterion in the form of linear matrix inequality (LMI) is derived with the aid of a Lyapunov function and a newly established fractional-order differential inequality. Then, the linear feedback pinning controller is employed for GMUB of the synchronization error system, which is equivalent to GMLQS of FMNNs. Since there exists the propagation delay between drive-response FMNNs and the synchronization error system can not be obtained straightforwardly, an identical equation about Caputo’s fractional derivatives is established to overcome this difficulty. Moreover, the detailed pinning control scheme design procedures for a prescribed GMLQS performance are presented, where the performance involves both ultimate error bound and transient behavior. In the control scheme, an auxiliary LMI and an optimization objective function are introduced, which can significantly reduce control cost. Finally, a numerical example is presented to show the feasibility of the control scheme. The results obtained in this paper have improved the existing criterion on quasi-synchronization of FMNNs, and will also provide a novel insight into pinning control of fractional-order nonlinear systems.