An overview of stability analysis and state estimation for memristive neural networks

Projective quasi-synchronization of coupled memristive neural networks with uncertainties and impulsive effect

The dynamic behavior of memristive neural networks (MNNs), including synchronization, effectively keeps the robotic stability against numerous uncertainties from the mimic of the human brain. However, it is challenging to perform projective quasi-synchronization of coupled MNNs with low-consumer control devices. This is partly because complete synchronization is difficult to realize under various projective factors and parameter mismatch. This article aims to investigate projective quasi-synchronization from the perspective of the controller. Here, two approaches are considered to find the event-triggered scheme for lag synchronization of coupled MNNs. In the first approach, the projective quasi-synchronization issue is formulated for coupled MNNs for the first time, where the networks are combined with time-varying delays and uncertainties under the constraints imposed by the frequency of controller updates within limited system communication resources. It is shown that our methods can avoid the Zeno-behavior under the newly determined triggered functions. In the second approach, following classical methods, a novel projective quasi-synchronization criterion that combines the nonlinear property of the memristor and the framework of Lyapunov-Krasovskii functional (LKF) is proposed. Simulation results indicate that the proposed two approaches are useful for coupled MNNs, and they have less control cost for different types of quasi-synchronization.

Nonlinear state estimation by Extended Parallelotope Set-Membership Filter

In this paper, we propose a state estimation method called the Extended Parallelotope Set-Membership Filter that provides a higher estimation accuracy than existing methods for discrete-time nonlinear systems. The Extended Parallelotope Set-Membership Filter is motivated by the fact that the iteration operations in existing methods generate much redundancy, and will deteriorate the accuracy of the state estimation. To account for this issue, an innovative parallelotope envelope method is proposed for the purpose of reducing the redundancy arising from the process of the noise envelope. In addition, a cofactor separation method is designed for nonlinear systems to obtain a tight envelope of the parallelotope set. Furthermore, we develop a novel parallelotope intersection method suitable for the parallelotope envelope to update the state set. The simulation results verified the effectiveness of the proposed method as well as its superiority over conventional methods in terms of both the maximum and average accuracies of the state estimation.

MANet: A two-stage deep learning method for classification of COVID-19 from Chest X-ray images

The early detection of infection is significant for the fight against the ongoing COVID-19 pandemic. Chest X-ray (CXR) imaging is an efficient screening technique via which lung infections can be detected. This paper aims to distinguish COVID-19 positive cases from the other four classes, including normal, tuberculosis (TB), bacterial pneumonia (BP), and viral pneumonia (VP), using CXR images. The existing COVID-19 classification researches have achieved some successes with deep learning techniques while sometimes lacking interpretability and generalization ability. Hence, we propose a two-stage classification method MANet to address these issues in computer-aided COVID-19 diagnosis. Particularly, a segmentation model predicts the masks for all CXR images to extract their lung regions at the first stage. A followed classification CNN at the second stage then classifies the segmented CXR images into five classes based only on the preserved lung regions. In this segment-based classification task, we propose the mask attention mechanism (MA) which uses the predicted masks at the first stage as spatial attention maps to adjust the features of the CNN at the second stage. The MA spatial attention maps for features calculate the percentage of masked pixels in their receptive fields, suppressing the feature values based on the overlapping rates between their receptive fields and the segmented lung regions. In evaluation, we segment out the lung regions of all CXR images through a UNet with ResNet backbone, and then perform classification on the segmented CXR images using four classic CNNs with or without MA, including ResNet34, ResNet50, VGG16, and Inceptionv3. The experimental results illustrate that the classification models with MA have higher classification accuracy, more stable training process, and better interpretability and generalization ability than those without MA. Among the evaluated classification models, ResNet50 with MA achieves the highest average test accuracy of 96.32% in three runs, and the highest one is 97.06%. Meanwhile, the attention heat maps visualized by Grad-CAM indicate that models with MA make more reliable predictions based on the pathological patterns in lung regions. This further presents the potential of MANet to provide clinicians with diagnosis assistance.

Dynamic event-based state estimation for delayed artificial neural networks with multiplicative noises: A gain-scheduled approach

This study is concerned with the state estimation issue for a kind of delayed artificial neural networks with multiplicative noises. The occurrence of the time delay is in a random way that is modeled by a Bernoulli distributed stochastic variable whose occurrence probability is time-varying and confined within a given interval. A gain-scheduled approach is proposed for the estimator design to accommodate the time-varying nature of the occurrence probability. For the sake of utilizing the communication resource as efficiently as possible, a dynamic event triggering mechanism is put forward to orchestrate the data delivery from the sensor to the estimator. Sufficient conditions are established to ensure that, in the simultaneous presence of the external noises, the randomly occurring time delays with time-varying occurrence probability as well as the dynamic event triggering communication protocol, the estimation error is exponentially ultimately bounded in the mean square. Moreover, the estimator gain matrices are explicitly calculated in terms of the solution to certain easy-to-solve matrix inequalities. Simulation examples are provided to show the validity of the proposed state estimation method.