查看更多>>摘要：Asynchronous spiking neural P (AsynSN P) systems are a class of distributed and parallel computational models working in a non-synchronized mode, inspired by the mechanism of information processing and communication underlies biological neurons by spikes, where at each computation step, any neuron with applicable rules is not obligatory to exe-cute an applicable rule. It remains an open problem whether AsynSN P systems using stan-dard spiking rules (the execution of the rule only produces a spike) are equivalent in power to Turing machines. In this work, the control mechanism of local synchronization at the rule level is introduced into AsynSN P systems. Namely, there are some given locally syn-chronous sets of rules; if a rule in such set is executed, then all applicable rules residing in the same set should be executed simultaneously. The computational power of AsynSN P systems with local synchronization of rules is examined. It is demonstrated that with local synchronization of rules, both general and unbounded AsynSN P systems are Turing uni-versal, whereas bounded AsynSN P systems can only characterize the family of semilinear sets of numbers, using standard spiking rules. These results demonstrate the great poten-tial for the local synchronization of rules to improve the computational capability of AsynSN P systems.(c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：Imbalanced dataset classification issue poses a major challenge on machine learning domain. Traditional supervised learning algorithms usually bias towards the majority class when handling imbalanced datasets, thus leading to poor classification results on the minority class. The learning task would become crucially difficult when there are overlap-ping and within-imbalance issues in imbalanced datasets, which are often the case and have been proven to severely deteriorate the classification performance relative to between-class imbalance. In this paper, we propose a novel SVDD boundary-based weighted oversampling approach (SVDDWSMOTE) for handling imbalanced and over -lapped data. The proposed approach first applies support vector data description (SVDD) model with greater penalty constant for the minority class than the majority class to gen-erate the class boundary, and then identifies those misclassified majority or few minority instances by the generated class boundary as potential overlapped or noisy ones and elim-inates them. To address the within-balance issues, we propose a weight assignment strat-egy based on densities and the distances to the SVDD class boundary, which facilitates simultaneously combating between-class and within-class imbalance issues caused by complicated distribution. In addition, such a strategy also favors generating more synthetic minority instances for borderline and sparser instances which are usually informative to the later learning tasks. Finally, oversampling is performed by the weighed SMOTE scheme based on SVDD boundary to not only counteract the within imbalance but also avoid the generation of any noisy or overlapped synthetic instance. Extensive comparison results on various datasets show that the proposed approach achieves statistically significant improvements in terms of different classification performance metrics relative to state -of-the-art ones.(c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：This paper proposes an all state constrained decentralized adaptive implicit inversion control scheme for a class of large scale nonlinear systems with unknown time delays and asymmetric saturated hysteresis. First, to address the states constrained problem, the asymmetric barrier Lyapunov function is introduced to keep the error surface within an appropriate range from the view of engineering practice to ensure the performance and safety, such as for attitude tracking of rigid spacecraft, for spacecraft approach and intersection. Second, the transmission delays between different subsystems are considered and approximated through the incorporation of the neural-network approximators and the finite coverage lemma. Third, a new hysteresis implicit inverse algorithm is designed to effectively mitigate asymmetric and saturated hysteresis nonlinearities. It should be noted that the implicit inverse implies that the analytical inverse of the asymmetric and saturated hysteresis is not required. Instead, the decoupling algorithms are designed to extract the actual control signal from the temporarily hysteretic control signal , which reduces the preliminary work of the control algorithm. Finally, all of the signals in the closed-loop system are proved to be semi-globally ultimately uniformly bounded and the tracking errors converge to an arbitrarily small residual set. The experimental results on two-machine excitation power systems in the hardware-in-loop system are presented to illustrate the effectiveness of the proposed scheme. (c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：The localization of hazardous sources (e.g. poisonous gas sources) is an important task regarding the security of human society. To find the unknown source in time, various autonomous source searching methods have mushroomed and been employed over the past decade. This paper designs a fresh source searching approach, namely particle clustering-deep Q-network, PC-DQN, which applies the deep reinforcement learning (DRL) techniques as a source searching approach for the first time. Specifically, the search process is formulated as the partially observable Markov decision process, then converted into the Markov decision process based on the belief state (represented by the particle fil-ter). PC-DQN leverages the density-based spatial clustering of applications with noise (DBSCAN) algorithm to extract the feature of belief state, and employ the deep Q-network (DQN) algorithm to find the optimal policy for the source searching task. Through the comparison with two baseline methods (i.e. RANDOM and Entrotaxis algo-rithm) under various experimental conditions, the viability of our proposed PC-DQN is tes-tified. Results explicitly reveal that the success rate of the PC-DQN maintains at a high level (beyond 99.6%) in all scenarios in this paper, and the mean search step shows evident supe-riority over baseline methods in most scenarios. Significantly, we also introduce the trans-fer learning concept to reuse the well-trained Q-network into new scenarios. These findings show important implications of the DRL-based approach as an alternative and more effective source searching approach.(c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：In color image processing, the objective of image completion is to restore missing entries from the incomplete observation image. Recent improvements have assisted in resolving the rank minimization issue, thereby promoting the realization of completion. This paper adopts the new quaternion matrix logarithmic norm to approximate rank in accordance with the quaternion matrix framework. Unlike the traditional matrix completion method, which handles RGB channels separately, the quaternion-based method avoids the destruction of the structure of images by placing the color image in a pure quaternion matrix. Furthermore, the logarithmic norm induces a more accurate rank surrogate. Based on the logarithmic norm, it is possible to exploit not only the factorization strategy but also the truncated technique, thus achieving successful image restoration. The alternating minimization framework renders it possible to optimize the two strategies, and mathematically detailed validation of the convergence analysis is provided. The experimental results demonstrate that the use of logarithmic surrogates in the quaternion domain is a superior strategy for solving the problem of color image completion. (c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：Recurrent neural networks have been reported as an effective approach to solve dynamic Lyapunov equations, which widely exist in various application fields. Considering that a bounded activation function should be imposed on recurrent neural networks to solve the dynamic Lyapunov equation in certain situations, a novel bounded recurrent neural network is defined in this paper. Following the definition, several bounded activation func-tions are proposed, and two of them are used to construct the bounded recurrent neural network for demonstration, where one activation function has a finite-time convergence property and the other achieves robustness against noise. Moreover, theoretical analyses provide rigorous and detailed proof of these superior properties. Finally, extensive simula-tion results, including comparative numerical simulations and two application examples, are demonstrated to verify the effectiveness and feasibility of the proposed bounded recur-rent neural network.(c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：Nonoverlapping sequential pattern mining (SPM) is a type of SPM with gap constraints that can mine valuable information in sequences. One of the disadvantages of nonoverlapping SPM is that any characters can match with gap constraints. Hence, there can be a significant difference between the trend of a pattern and those of its occurrences. To tackle this issue, we propose nonoverlapping weak-gap sequential pattern (NWP) mining, where characters are divided into two types: weak and strong. This allows discovering frequent patterns more accurately by limiting the gap constraints to match only weak characters. To discover NWPs, we propose NMP-Miner which involves two key steps: support calculation and candidate pattern generation. To efficiently calculate the support of candidate patterns, depth-first search and backtracking strategies based on a simplified Nettree structure are adopted, which effectively reduce the time and space complexities of the algorithm. Moreover, a pattern join approach is applied to effectively reduce the number of candidate patterns. The experimental results show that NWP-Miner is more efficient than other competitive algorithms. More importantly, the case study of time series shows that NWP-Miner can effectively filter out noise patterns and discover more meaningful patterns. Algorithms and datasets can be downloaded from https://github.com/wuc567/ Pattern-Mining/tree/master/NWP-Miner.(c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：When addressing analysis and prediction problems in a specific domain based on big data processing, the following problems often arise: only relationships between features in the domain itself are considered, and existing methods are not effective for training models on small sample data sets. The traditional approach usually obtains the relationships between single-domain features. Analysis and forecasting in the problem domain alone can quickly achieve good accuracy, but due to the limitations of the analysis domain, it becomes increasingly difficult to further improve the prediction accuracy. This paper proposes a novel data analysis approach compatible with small sample sets called multidomain data depth analysis (MODE). In contrast to traditional approaches, MODE emphasizes multido-main data and considers the relationships among feature domains in the original data. The features in each domain are orthogonally extracted, and feature dimensions are expanded in accordance with the characteristics of small data sets. A better prediction model can be obtained by using the expanded and strengthened features. We apply this approach to real big data from the field of sociology to predict annual income based on census data in exper-iments. The experimental results show that MODE offers a better prediction effect based on small multidomain samples.(c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：In this paper, a denoising temporal convolutional recurrent autoencoder (DTCRAE) is proposed to improve the performance of the temporal convolutional network (TCN) on time series classification (TSC). The DTCRAE consists of a TCN encoder and a Gated Recurrent Unit (GRU) decoder. Training the DTCRAE for TSC includes two phases, an unsupervised pre-training phase based on a DTCRAE and a supervised training phase for developing a TCN classifier. Computational studies are conducted to prove the effectiveness of DTCRAEs for TSC based on three datasets, the Sequential MNIST, Permuted MNIST, and Sequential CIFAR-10. Computational results demonstrate that the pre-trained DTCRAE provides a better initial structure for a TCN classifier, in terms of its higher precisions, recalls, F1-scores, and accuracies. The sensitivity analysis on the validation set shows that the pre trained DTCRAE is robust to changes of the batch size, noisy rate, and dropout rate. DTCRAEs offer best TSC accuracies on two of three datasets and an accuracy comparable to the best one on another dataset by benchmarking against a number of state-of-the-art algorithms. Results verify the advantage of applying DTCRAEs to enhance the TSC performance of the TCN.(c) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：Smartphones enable to collect large data streams about phone calls that, once combined with Computational Intelligence techniques, bring great potential for improving the monitoring of patients with mental illnesses. However, the acoustic data streams recorded in uncontrolled environments are dynamically changing due to various sources of uncertainty. In addition, such acoustic data are usually difficult to interpret by psychiatrists. Within this study, we propose an approach based on Linguistic Summaries with Fuzzy Clustering (LS-FC) aiming at the development of human-consistent and easily interpretable summaries about relations between acoustic data and mental state of a patient affected by Bipolar Disorder, e.g., Most calls in the state of hypomania have low loudness compared to the state of euthymia [T = 1]. To capture the dynamics of acoustic data streams, we apply a dynamic incremental semi-supervised fuzzy clustering that synthesizes data into clusters. These clusters are represented by prototypes which are used for the construction of the membership functions describing linguistic terms e.g., low loudness, and then, linguistic summaries. The main contribution of this paper is the incorporation of information about clusters' prototypes in the generation of linguistic summaries. The primary goal of this research is explainability. The semi-supervised learning algorithm is used mainly for deriving clusters and building improved linguistic summaries. Numerical results indicate that linguistic summaries provide intuitive and clear information about voice features in a patient's affective state and they are consistent with clinical observation. In particular, during most calls in hypomania/mania both the quality of the patient's voice and the dynamics of change in the spectrum signal reflected in spectral flux are low compared to euthymia. The proposed approach enables to summarize large data streams into meaningful descriptions that, although relatively simple, offer information granules that are very intuitive for clinicians and are promising to support the smartphone-based monitoring of bipolar disorder patients to inform about the potential change of mental state. (c) 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BYNC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
NSTL
Elsevier