This study proposes a novel and cohesive framework to address stochastic finite-time consensus (FTC) problems, with the following main contributions: (i) We first introduce the original stochastic delay systems and, based on this, analyze the effects of Lévy noise, actuator faults, and Markov switching. Both leaderless and leader-follower topologies are considered, and a new control algorithm is proposed to investigate the fault-tolerant control problem under the influence of communication delays and Markov switching dynamics. (ii) To ensure that the states converge to a bounded compact set, the convergence analysis uses strong mathematical techniques, such as stopping time theory and the evolution of finite-time stochastic theory, to achieve mean-square and almost certain consensus. (iii) An important aspect of this study is the consideration of Markov-switching actuator faults, where fault occurrence and recovery evolve randomly according to a Markov process, introducing additional stochastic uncertainties into the system dynamics. Additionally, two numerical examples are provided to validate the correctness of the theoretical results.

This article studies the distributed coordination control problem of the nonlinear stochastic multiagent system, which involves multiple inequality constraints, as well as random disturbances. The communication network is modeled as an undirected and connected graph, where each node has a cost function that is considered to be strongly convex. First, an algorithm is developed that utilizes state feedback and projection operations, along with auxiliary variables, to precisely estimate the optimal state and its derivatives in a distributed manner. Furthermore, the decision variable is subject to several inequality constraints, and there are no restrictions on the initial value. Taking into account the impact of nonlinearity resulting from drift coefficients, diffusion coefficients, and external disturbances on the system’s stability, the existing works cannot be directly applied to this study. Through the utilization of the${\rm It}\hat{\rm o}$formula and convex analysis, a novel analytical approach has been demonstrated to establish the asymptotic convergence of decision variable to the optimal value in mean square. This method distinguishes itself from the conventional convergence analysis techniques. Finally, the algorithm is utilized for the allocation of energy resources, resulting in the attainment of the optimal regulating method for power resource allocation.

Cross-domain recommendation aims to leverage heterogeneous information to transfers knowledge from a data-sufficient domain (source domain) to a data-scarce domain (target domain). Existing approaches mainly focus on learning single-domain user preferences and then employ a transferring module to obtain cross-domain user preferences, but ignore the modeling of users' domain specific preferences on items. We argue that incorporating domain-specific preferences from the source domain will introduce irrelevant information that fails to the target domain. Additionally, directly combining domain-shared and domain-specific information may hinder the target domain's performance. To this end, we propose C^2DR, a novel approach that disentangles domain-shared and domain-specific preferences from a causal perspective. Specifically, we formulate a causal graph to capture the critical causal relationships based on the underlying recommendation process, explicitly identifying domain-shared and domain-specific information as causal irrelevant variables. Then, we introduce disentanglement regularization terms to learn distinct representations of the causal variables that obey the independence constraints in the causal graph. Remarkably, our proposed method enables effective intervention and transfer of domain-shared information, thereby improving the robustness of the recommendation model. We evaluate the efficacy of C^2DR through extensive experiments on three real-world datasets, demonstrating significant improvements over state-of-the-art baselines.

The cross-domain recommendation (CDR) model addresses challenges such as data sparsity, the long tail distribution of user-item interactions, and the cold start of items or users. However, solely transferring domain-shared knowledge based on the co-occurrence patterns, without considering user preferences, leads to negative transfer in CDR. To overcome these limitations, we propose an advanced deep learning CDR model called the Domain Adversarial Deep Interest Network (DADIN) aims to facilitate smooth knowledge transfer from the source domain to the target domain and effectively alleviate negative transfer. Firstly, the joint distribution alignment of user preference in DADIN is realized by introducing a skip-connection-based domain agnostic layer, and then the domain classifier is artificially designed to distinguish between the information coming from the source domain or the target domain. Additionally, DADIN combines prediction loss, global domain confusion loss, and intra-class domain confusion losses through the Min-Max game and gradient reverse layer to achieve collaborative optimization. Two real-world experiments show the area under curve (AUC) of DADIN is 0.78 on the Huawei dataset, and it outperforms its competitors by 0.71% on the Amazon dataset, showcasing its state-of-the-art performance. Moreover, our ablation studies further demonstrate that domain adversarial technique increases the AUC by 2.34% on the Huawei dataset and 16.67% on the Amazon dataset, respectively.

Recommendation system algorithm based on multi-task learning (MTL) is the major method for Internet operators to understand users and predict their behaviors in the multi-behavior scenario of platform. Task correlation is an important consideration of MTL goals, traditional models use shared-bottom models and gating experts to realize shared representation learning and information differentiation. However, The relationship between real-world tasks is often more complex than existing methods do not handle properly sharing information. In this paper, we propose an Different Expression Parallel Heterogeneous Network (DEPHN) to model multiple tasks simultaneously. DEPHN constructs the experts at the bottom of the model by using different feature interaction methods to improve the generalization ability of the shared information flow. In view of the model's differentiating ability for different task information flows, DEPHN uses feature explicit mapping and virtual gradient coefficient for expert gating during the training process, and adaptively adjusts the learning intensity of the gated unit by considering the difference of gating values and task correlation. Extensive experiments on artificial and real-world datasets demonstrate that our proposed method can capture task correlation in complex situations and achieve better performance than baseline models.

There are two fundamental problems in applying deep learning/machine learning methods to disease classification tasks, one is the insufficient number and poor quality of training samples; another one is how to effectively fuse multiple source features and thus train robust classification models. To address these problems, inspired by the process of human learning knowledge, we propose the Feature-aware Fusion Correlation Neural Network (FaFCNN), which introduces a feature-aware interaction module and a feature alignment module based on domain adversarial learning. This is a general framework for disease classification, and FaFCNN improves the way existing methods obtain sample correlation features. The experimental results show that training using augmented features obtained by pre-training gradient boosting decision tree yields more performance gains than random-forest based methods. On the low-quality dataset with a large amount of missing data in our setup, FaFCNN obtains a consistently optimal performance compared to competitive baselines. In addition, extensive experiments demonstrate the robustness of the proposed method and the effectiveness of each component of the model.

The main purpose of this paper was to develop a deep-learning method for the diagnosis of different chronic middle ear diseases, including middle ear cholesteatoma and chronic suppurative otitis media, based on computed tomography (CT) images of the middle ear. The origin of the dataset was the CT scans of 499 patients, which included both ears and selected by specialized otologists. The final dataset was constructed from 973 ears, which labeled by a professional otolaryngologist and classified into 3 conditions: MEC, CSOM and normal. The diagnostic framework, called the “Middle Ear Structure Identification Classifier”(MESIC), was consisted of two deep-learning networks with dissimilar functions: a “region of interest” area search network for extracting the special image of the middle ear structure and a classification network for finishing the diagnosis. The area under the curve (AUC), which means receiver operating characteristic curve (ROC), reflects the robustness of the algorithm by comparing its sorting effectiveness. According to simulation experiments, we chose Visual Geometry Group 16 (VGG-16) as the model’s backbone. In our framework, the ROI search part exhibited an AUC of 0.99 on the right and 0.98 on the left. The classification part exhibited an average AUC of 0.96 for both sides based on VGG-16. The average precision (90.1%), recall (85.4%) and F1-score (87.2%) show the effectiveness of framework. This paper presents a deep-learning framework to automatically diagnose cholesteatoma and CSOM. The results show that MESIC can effectively and quickly classify these two common diseases through CT images, which can ameliorate the pressure of professional doctors and the practical problems of the lack of professional doctors in rural areas.

It is well known that single hidden layer feedforward networks with radial basis function (RBF) kernels are universal approximators when all the parameters of the networks are obtained through all kinds of algorithms. However, as observed in most neural network implementations, tuning all the parameters of the network may cause learning complicated, poor generalization, overtraining and unstable. Unlike conventional neural network theories, this brief gives a constructive proof for the fact that a decay RBF neural network with n+1 hidden neurons can interpolate n+1 multivariate samples with zero error. Then we prove that the given decay RBFs can uniformly approximate any continuous multivariate functions with arbitrary precision without training. The faster convergence and better generalization performance than conventional RBF algorithm, BP algorithm, extreme learning machine and support vector machines are shown by means of two numerical experiments.