Image segmentation plays a key role in many image-guided clinical applications and deep learning technology has proven effective for this task when sufficient labeled images are available. However, it is very time-consuming and labor-intensive to obtain adequate pixel-level image labels. To alleviate the scarcity of labeled images, we propose a novel semi-supervised segmentation method based on the available uncertainty-aware mean teacher (UAMT) framework by introducing two different strategies, i.e., selective self-ensembling (SSE) and boundary uncertainty suppression (BUS). The SSE dynamically selects multiple best student models across different training steps to update the teacher model's weights, while the BUS reduces boundary segmentation errors and improves the quantitative potential of loss functions through a unique uncertainty estimation function. With the two strategies, our proposed method was able to obtain promising segmentation performance with limited labeled images and abundant unlabeled ones. We trained and validated our proposed method by segmenting multiple objects from three public datasets (i.e., PROMISE, REFUGE, and RETA). Extensive experiments showed that our proposed method achieved better segmentation performance than the UAMT, along with the average Dice score (DSC) of 0.7990 for three different objects, and can compete with several existing semi-supervised methods (i.e., HCMT, SASSNet, and DTC). • A novel semi-supervised learning method was developed for accurate image segmentation. • A performance-driven exponential moving average (pEMA) was proposed for semi-supervised learning. • A unique exponential function was proposed for uncertainty estimation. • Extensive segmentation experiments showed the advantage of the developed method.

Equipping satellites with computing resources to ensure efficient mission completion has become a pivotal trend in multi-layer satellite networks (MLSNs). The uneven spatial distribution of missions and computing resources across satellites necessitates advanced scheduling of communication and computing resources through satellite collaboration. However, the intricate interactions between communication and computing resources, the dynamic mission arrivals and computing resources, and the difficulty of collaboration across different layers in MLSNs present significant challenges for effective scheduling. This paper proposes a collaborative scheduling framework for low Earth orbit (LEO) and medium Earth orbit (MEO) satellites to support communication-computing integration in the MLSN. To adapt to network dynamics, we introduce a federated aggregation matrix and propose an intelligent MEO-LEO collaborative scheduling algorithm that optimizes the decision-making process under uncertain mission arrivals. Additionally, we design a distributed LEO-LEO collaborative scheduling algorithm that leverages the synergy between inter-satellite communication and computing resources to enhance scheduling capabilities and create communication-computing resource chains that meet mission requirements. Extensive simulations demonstrate that our proposed collaborative scheduling framework significantly enhances the scheduling capability of the MLSN.

As one of main active learning methods, zero-shot target recognition with synthetic aperture radar (SAR) data has received considerable attention in recent years. Its goal is to distinguish new targets from the known classes without requiring additional training data. Existing zero-shot learning (ZSL) methods perform well on optical targets, but they fail to recognize zero-shot SAR targets. Due to the strong similarity of SAR targets between different classes, the ZSL task usually suffers from the distribution concentration problem. To tackle this problem, a Dual Branch Auto-Encoder (DBAE) network is proposed in this letter. DBAE effectively alleviates the distribution concentration problem by adding a classification assistance net. Its dual branch structure is specially designed for further improving the intra-class similarity and inter-class dissimilarity of SAR targets in the embedding space. By training with the defined hybrid loss function, DBAE automatically builds a stable embedding space. Extensive experiments on the public data set of Moving and Stationary Target Acquisition and Recognition (MSTAR) show that DBAE is of rational design and provides better or comparable ZSL recognition results.

The expansion of satellite applications induces the formation of the multi-domain satellite system (MDSS) containing multiple domains with specific applications such as earth resource remote sensing and the Internet of remote things. Resource management is pivotal in enhancing the scheduling capability of the MDSS. However, this is challenging since the dynamic buffer space and communication opportunity, as well as the uncertain data traffic, exacerbate the difficulty of matching satellite resources with data traffic. Moreover, the coexistence of resource competition and collaboration across domains aggravates the dilemma of cross-domain collaboration. In this paper, we propose a hierarchical cross-domain collaborative resource management framework that can flexibly allocate the mission data through local intra-domain and global cross-domain scheduling. Then, to match the uncertain demands of missions with dynamic and limited resources, we propose a multi-agent reinforcement learning-based resource management method to guide collaboration for multi-satellite data carry-forward in a domain. Further, considering resource competition and collaboration in MDSS, we propose a domain-satellite nested matching game data scheduling algorithm to achieve pair-wise stable collaboration of cross-domain satellites. The simulation results indicate that the proposed algorithm improves the amount of offloaded data by 64.4% and 12.7% compared to the non-collaborative and the non-cross-domain schemes, respectively.

In real-world recognition tasks, it is difficult to collect training samples for all types of synthetic aperture radar (SAR) targets. The existing zero-shot learning (ZSL) methods use visual features and label semantics simultaneously to ensure a high recognition rate of unknown targets. However, meaningless information could influence the stability of the reference space. The negative impact will further drop the recognition rate of the ZSL methods. To overcome this problem, a new ZSL method is proposed by only adopting reflection similarity to recognize unknown targets. The proposed method adopts a triple-part network, in which the constructor-net and the supervisor-net jointly construct the reference space. It makes the space have a good representation and distinction. The third part of the triple-part network, as an interpreter-net, provides the uniformity of the target interpretation. The experimental results on moving and stationary target acquisition and recognition (MSTAR) data set show that the reference space with good quality can be obtained by discarding meaningless information. Especially, when the samples of known targets have slightly changed, the reference space still keeps good stability. Compared with the traditional ZSL methods, the recognition reliability of the proposed method on zero-shot targets is improved.

An efficient resource allocation algorithm plays a pivotal role in the performance improvement of space information networks (SIN). The dynamic of resources and mission requirements in the network has a great influence on resource allocation. Guiding rapid satellite resource allocation to adapt to dynamic changes of the network by discovering the similarity in the structure of changing missions is a key technology to improve SIN performance. In this paper, we firstly formulate satellite resource allocation as a problem aiming to maximize the total network benefits. Then, we propose a satellite resource allocation algorithm based on Hopfield to solve resource allocation problems in SIN. To accommodate the dynamics of satellite network missions and quickly solve resource allocation problems, we further propose a satellite resource allocation algorithm based on transfer learning to learn the node selection strategy in the Hopfield network. Specifically, we use a small number of additional training samples to better adapt to the changes in mission structure. Simulation results validate that the proposed method can achieve high performance while reducing computational complexity.

The massive increase of spam is posing a very serious threat to email and SMS, which have become an important means of communication. Not only do spams annoy users, but they also become a security threat. Machine learning techniques have been widely used for spam detection. In this paper, we propose another form of deep learning, a linguistic attribute hierarchy, embedded with linguistic decision trees, for spam detection, and examine the effect of semantic attributes on the spam detection, represented by the linguistic attribute hierarchy. A case study on the SMS message database from the UCI machine learning repository has shown that a linguistic attribute hierarchy embedded with linguistic decision trees provides a transparent approach to in-depth analysing attribute impact on spam detection. This approach can not only efficiently tackle `curse of dimensionality' in spam detection with massive attributes, but also improve the performance of spam detection when the semantic attributes are constructed to a proper hierarchy.

Internet of Things (IoT) has given rise to the fourth industrial revolution (Industrie 4.0), and it brings great benefits by connecting people, processes and data. However, cybersecurity has become a critical challenge in the IoT enabled cyber physical systems, from connected supply chain, Big Data produced by huge amount of IoT devices, to industry control systems. Evolutionary computation combining with other computational intelligence will play an important role for cybersecurity, such as artificial immune mechanism for IoT security architecture, data mining/fusion in IoT enabled cyber physical systems, and data driven cybersecurity. This paper provides an overview of security challenges in IoT enabled cyber-physical systems and what evolutionary computation and other computational intelligence technology could contribute for the challenges. The overview could provide clues and guidance for research in IoT security with computational intelligence.

The massive increase of spam is posing a very serious threat to email and SMS, which have become an important means of communication. Not only do spams annoy users, but they also become a security threat. Machine learning techniques have been widely used for spam detection. Email spams can be detected through detecting senders' behaviour, the contents of an email, subject and source address, etc, while SMS spam detection usually is based on the tokens or features of messages due to short content. However, a comprehensive analysis of email/SMS content may provide cures for users to aware of email/SMS spams. We cannot completely depend on automatic tools to identify all spams. In this paper, we propose an analysis approach based on information entropy and incremental learning to see how various features affect the performance of an RBF-based SVM spam detector, so that to increase our awareness of a spam by sensing the features of a spam. The experiments were carried out on the spambase and SMSSpemCollection databases in UCI machine learning repository. The results show that some features have significant impacts on spam detection, of which users should be aware, and there exists a feature space that achieves Pareto efficiency in True Positive Rate and True Negative Rate.

A CMAC (Cerebellar Model Articulation Controller) is a kind of feed-forward neural networks (FFNNs), but the feature of fast learning makes it different from classic FFNNs. A CMAC has a single linear trainable layer, but due to the input information is distributed in a hypercube grid, it is suitable for modeling any non-linear relationship. It has been proved that a linguistic CMAC (LCMAC) based on the Label Semantics can represent the rules that a linguistic decision tree (LDT) does when they are used to map the relationship between inputs and outputs. In order to overcome the `curse of dimensionality' of an LCMAC in memory use for multi-attribute decision making, a cascade of LCMACs is proposed, and the linguistic interpretation of a cascade of LCMACs is investigated to break the convention of neural networks as a black box. An algorithm for training a cascade of LCMACs is developed. The proposed cascade of LCMACs has a great reduction in the use of memory units, and as a decision maker, it can achieve better performance in accuracy and F1-score than the cascade of LDTs does for those data sets with non-linear relations between inputs and outputs.

Machine learning enables the creation of a nonlinear mapping that describes robot-environment interaction, whereas computing linguistics make the interaction transparent. In this paper, we develop a novel application of a linguistic decision tree for a robot route learning problem by dynamically deciding the robot's behavior, which is decomposed into atomic actions in the context of a specified task. We examine the real-time performance of training and control of a linguistic decision tree, and explore the possibility of training a machine learning model in an adaptive system without dual CPUs for parallelization of training and control. A quantified evaluation approach is proposed, and a score is defined for the evaluation of a model's robustness regarding the quality of training data. Compared with the nonlinear system identification nonlinear auto-regressive moving average with eXogeneous inputs model structure with offline parameter estimation, the linguistic decision tree model with online linguistic ID3 learning achieves much better performance, robustness, and reliability.

ABSTRACT Before the development of a large‐scale wireless sensor network (WSN) infrastructure, it is necessary to create a model to evaluate the lifespan of the infrastructure, the system performance and the cost so that the best design solution can be obtained. Energy consumption is an important factor that influences the lifespan of WSNs. One of the ways to extend the lifespan of WSNs is to design wireless sensor nodes with low power consumption. This involves component selection and the optimisation of hardware architecture, monitoring software system and protocols to satisfy the requirements of the particular applications. This paper proposes a comprehensive model to describe the workflow of a wireless sensor node. Parameter setup and energy consumption calculation are demonstrated through the model simulation. It provides a mathematical approach to dynamically evaluate the energy consumption of a sensor node. This will benefit the development of wireless sensor nodes based on microprocessors with limited computational capability. Therefore, the model can be applied in dynamic power management systems for wireless sensor nodes or in wireless communication protocols with energy awareness, in particular, for WSNs with self‐organisation. More importantly, the generalisation of the model may be employed as a standard paradigm for the development of wireless sensor node with energy awareness. Copyright © 2012 John Wiley & Sons, Ltd.

Linguistic Decision Trees based on label semantics have been used as a classifier or predictor in many areas. A linguistic decision tree presents information propagation from input attributes to a goal variable based on transparent linguistic rules. The relationship between input attributes and the goal variable is often highly nonlinear. Cerebellar Model Articulation Controller (CMAC) belongs to the family of feed-forward networks with a single linear trainable layer. A CMAC has the feature of fast learning, and is suitable for modeling any non-linear relationship. Combining label semantics and an original CMAC, a linguistic CMAC based on Mass Assignment on labels is proposed to map the relationship between the attributes and the goal variable. The proposed LCMAC model is functionally equivalent to a linguistic decision tree, and takes the advantage of fast local training of the original CMAC and the advantage of transparency of a linguistic decision tree.

A wireless ad hoc sensor network consists of a number of sensors spreading across a geographical area. The performance of the network suffers as the number of nodes grows, and a large sensor network quickly becomes difficult to manage. Thus, it is essential that the network be able to self-organize. Clustering is an efficient approach to simplify the network structure and to alleviate the scalability problem. One method to create clusters is to use weakly connected dominating sets (WCDSs). Finding the minimum WCDS in an arbitrary graph is an NP-complete problem. We propose a neural network model to find the minimum WCDS in a wireless sensor network. We present a directed convergence algorithm. The new algorithm outperforms the normal convergence algorithm both in efficiency and in the quality of solutions. Moreover, it is shown that the neural network is robust. We investigate the scalability of the neural network model by testing it on a range of sized graphs and on a range of transmission radii. Compared with Guha and Khuller's centralized algorithm, the proposed neural network with directed convergency achieves better results when the transmission radius is short, and equal performance when the transmission radius becomes larger. The parallel version of the neural network model takes time O(d), where d is the maximal degree in the graph corresponding to the sensor network, while the centralized algorithm takes O(n2). We also investigate the effect of the transmission radius on the size of WCDS. The results show that it is important to select a suitable transmission radius to make the network stable and to extend the lifespan of the network. The proposed model can be used on sink nodes in sensor networks, so that a sink node can inform the nodes to be a coordinator (clusterhead) in the WCDS obtained by the algorithm. Thus, the message overhead is O(M), where M is the size of the WCDS.

We investigate the propagation of label informa tion for multi-attribute decision maldng problems downwards a linguistic attribute hierarchy, which represents the complex and often imprecise functional relationships between low level attributes or measurements and high-level decision or classification variables. The downward propagation algorithm identifies the branches in linguistic decision trees for which the probability of a high-level goal exceeds a given threshold. The sensitivity of the method to this threshold is then reduced by integrating with respect to a probability distribution on high threshold values.

We propose label semantics as an integrated representation framework for probabilistic uncertainty and fuzziness in multiple-attribute decision making problems. Linguistic attribute hierarchies are then introduced as a means of modelling the complex and often imprecise functional relationships between low-level attributes or measurements and high-level decision or classification variables. Within this framework we introduce linguistic decision trees as a tool for information aggregation in multi-attribute decision problems and describe the process of information propagation through a hierarchy of linked decision trees. In addition, we consider the ranking of different alternatives or examples based on their linguistic descriptions of a high-level utility variable. Finally, we discuss how linguistic decision trees embedded in attribute hierarchies can be learnt from data.

We propose label semantics as an integrated representation framework for probabilistic uncertainty and fuzziness in multiple-attribute decision making problems. Linguistic attribute hierarchies are then introduced as a means of modelling the complex and often imprecise functional relationships between low-level attributes or measurements and high-level decision or classification variables.

The simplest graph drawing method is that of putting the vertices of a graph on a line and drawing the edges as half-circles either above or below the line. Such drawings are called two-page book drawings. The smallest number of crossings over all two-page drawings of a graph G is called the two-page crossing number of G. Cimikowski and Shope have solved the two-page crossing number problem for an n-vertex and m-edge graph by using a Hopfield network with 2 m neurons. We present here an improved Hopfield model with m neurons. The new model achieves much better performance in the quality of solutions and is more efficient than the model of Cimikowski and Shope for all graphs tested. The parallel time complexity of the algorithm, without considering the crossing number calculations, is O(m) for the new Hopfield model with m processors clearly outperforming the previous algorithm.