查看更多>>摘要：An optimization spiking neural P system (OSNPS) aims to obtain the approximate solutions of combinatorial optimization problems without the aid of evolutionary operators of evo-lutionary algorithms or swarm intelligence algorithms. To develop the promising and sig-nificant research direction, this paper proposes a distributed adaptive optimization spiking neural P system (DAOSNPS) with a distributed population structure and a new adaptive learning rate considering population diversity. Extensive experiments on knapsack prob-lems show that DAOSNPS gains much better solutions than OSNPS, adaptive optimization spiking neural P system, genetic quantum algorithm and novel quantum evolutionary algo-rithm. Population diversity and convergence analysis indicate that DAOSNPS achieves a better balance between exploration and exploitation than OSNPS and AOSNPS. (c) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：Since cloud computing has been developing as a successful business and growing interest of organizations to make use of the environment, the volume of provided services is highly demanding. Cloud computing environment has high maintenance costs, therefore, the resources are limited; due to rapid progress and the highly demanded volume of utilization of resources, ascertaining an appropriate approach for controlling and managing resources has become essential in this area. Therefore, an algorithm must be designed to perform better in all QoS. A significant factor in current exploratory methods is that each method owns distinctive features and perhaps does not estimate all QoSes. For example, one speci-fic method has low energy consumption but does not consider SLA violation. Concerning the aims of the present study, the granular model is designed which uses various criteria involved in computations; in practice, by entering the values given in the past and discov-ering the relationships among them, the most desirable classification is produced. The membership function and the inference rule are derived from the data provided by the actual workload. Experimental analysis of the results showed that this method works out better than all other related methods and avails consistent performance in all QoS criteria.(c) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：Evolving classifiers and especially evolving fuzzy classifiers have been established as a prominent technique for addressing the recent demands in building classifiers in an incremental open-loop manner, e.g. for the purpose of processing data streams online. So far, the focus lies on classifiers which are obtained based on the input and/or feedback in the form of target labels provided by a single user/expert. In this paper, we propose three variants of evolving multi-user fuzzy classifier systems (EFCS-MU), where multiple users may provide their label feedback: i) ensembled single-user classifiers system, which allows a separate classifier training per user and embeds an advanced aggregation strategy (-> ensembling on a model level), ii) consensus all-user classifier system, where a joint classifier is established for all users based on consensus labelings (-> ensembling on a label level), iii) shift-work all -user classifier system, where a joint classifier is established for all users based on the classical shift-work concept. The classifiers are incrementally evolved by a single-pass learning approach embedding the autonomous evolution of new rules on demand; it integrates an unsupervised evolving clustering technique for rule partitioning, thus the same partition is established in all single-user classifiers, only the consequents in the form of class confidence vectors typically differ among the users due to their different labelings. This offers direct explainability of the varying users' annotation behaviors. The possible different experience levels of the users in relation to the process behind and possible ambiguities among the provided users' labels are handled by the proper integration of uncertainty levels into the update of the classifier(s). Furthermore, a concept is presented as to how to adequately integrate possibly available expert rules for a particular newly (on-the-fly) arising class (or in advance for several classes). Finally, an on-line active learning (oAL) strategy is demonstrated, to select only the most important samples to be labelled and thus reduce users' labeling costs, ensuring economically practicable applicability. The approach was successfully evaluated on two real-world application scenarios, one stemming for a visual inspection scenario, where four users check the quality of the imprint of compact discs and are affected by different experience levels, and one from a nursery school employment ranking application, where a new class was introduced later. The results provide insights into the performance behavior of the three different multi-user classifier variants under different circumstances (with and without expert rules, uncertainty integration, different labelling budgets etc.), including comparisons based on on-line accuracy trends versus the economy of the labeling effort. (c) 2022 The Author. Published by Elsevier Inc.
查看更多>>摘要：Multi-label datasets often contain label information with missing values and recovering them is a non-trivial challenge. Several methods augment the observed label matrix by constructing auxiliary labels and learning high order label correlations. Some other techniques exploit the low rank of the label matrix to capture a mix of label correlations. Both these approaches rely on label correlations, however in different ways. In this paper, we propose a unified framework that captures the label correlations utilizing both auxiliary label matrix and the low rank constraints on estimated labels. Our model also enforces maximal separation among different label subspaces for better label differentiation. The proposed method captures local and global correlations using Low Rank label subspace transformation for Multi-label learning with Missing Labels (LRMML). The model considers an auxiliary label matrix which facilitates the missing label information recovery. Low rank on predictions ensures that local label structures are captured and the maximal inter-label subspace separation helps identify discriminatory label correlations. The proposed method builds a multi-label classification model by solving a multivariate difference of convex objective function using surrogate optimization technique and alternating minimization. Empirical results on several benchmark datasets validate the effectiveness of the proposed method against state-of-the-art multi-label learning approaches.(c) 2022 Elsevier Inc. All rights reserved.
Palanimuthu, KumarasamyKim, Han SolJoo, Young Hoon
20页
查看更多>>摘要：This study is mainly concerned with fuzzy integral sliding mode control (FISMC) design for double-fed induction generator (DFIG)-based wind energy system (WES) under membership function-dependent H-infinity approach. To do this, a nonlinear DFIG based WES is considered with exogenous load torque variation. Then, by utilizing Takagi-Sugeno fuzzy approach, the proposed nonlinear system can be expressed as a sum of local sub-models with the help of IF-THEN rules. A suitable FISMC is designed with a reaching law condition for the proposed DFIG-based WES with the disturbances. A new fuzzy-based Lyapunov function is constructed, and the H-infinity performance index is calculated for external disturbances under membership function-dependent H-infinity approach. The obtained performance index of membership function-dependent H-infinity approach can be improved 10% than the conventional H-infinity approach. Finally, the simulation results are given the better stable performance of the DFIG-based WES under the designed FISMC, and the comparison results verify the efficacy of the proposed method. (c) 2022 Published by Elsevier Inc.
查看更多>>摘要：Overlapping communities are ubiquitous in real-world systems. For overlapping community detection, local expansion methods excel in scalability and efficiency yet have poor tolerance to low-quality seeds and communities. Based on our previous work, we introduce a more robust local-expansion-based overlapping community detection algorithm, named CEO, performing Construction, Expansion and Optimization sub-processes. To solve the poor fault tolerance problem, CEO discards low-quality seeds and communities in each sub-process based on optimizing node memberships. CEO was compared to thirteen noted algorithms by examining the performance on five groups of artificial networks and sixteen real-world networks with ground-truth communities. Experimental results showed CEO performs the best in identifying overlapping communities, which verifies the effectiveness of discarding low-quality seeds and communities in solving the poor fault tolerance problem.(c) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：Linear regression is an ordinary machine learning algorithm that models the relation between the input values and the output ones with underlying linear functions. Giacomelli et al. (ACNS 2018) proposed the first system training the linear regression model over the rational num-bers using only linearly homomorphic encryption. However, we find their system model is not applicable. A third authority generates the public key and secret key, which are used to encrypt and decrypt all the data sets. Then the privacy of data sets is in the risk of leakage even if the third authority is assumed to have no access to encrypted data sets. In this paper, we improve the system model in order to design a more practical linear regression algorithm over the rational numbers from the view of security. Concretely, every data owner generates his own public key and secret key, independent on a third authority. An improved multi-key fully homomorphic encryption over complex numbers is utilized to construct our linear regression algorithm with a preprocessing phase, which can directly encrypt rational numbers, support computations over ciphertexts under multi keys and obviate the rational reconstruction tech-nique as Giacomelli et al.. Furthermore, performance analyses demonstrate that our algorithm is more feasible and practical. (c) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：This work investigates the self-triggered control issue for interval type-2 (IT2) fuzzy sys-tems. The transmission of the system state is determined via a self-triggered scheme. Instead of continuously monitoring the prespecified triggering condition, the self-triggered scheme can calculate the next triggering instant by the latest triggered state information. Then, by means of the triggered state, a self-triggered IT2 fuzzy controller is designed, whose fuzzy rules and membership functions (MFs) can be flexibly selected inde-pendent of the controlled fuzzy system. Under the constraint of imperfectly premise matching, the relationship between the triggered state and the system state is explicitly established, and the boundaries of MFs in the local operation region are given. Meanwhile, the local boundary information is utilized to obtain the relaxed feasible condi-tions. Finally, the simulation examples are give and the comparison with other triggered schemes are also made.CO 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：Multi-view discriminant analysis (MvDA) is a successful method to learn a single discriminant common space of multiple views. However, MvDA may encounter the robustness issue theoretically because of using F-norm as the metric. In this paper, a robust multi view discriminant analysis with view-consistency is proposed by employing L-1-norm as the metric, called as L-1-MvDA-VC. where both inter-view and intra-view distances are characterized by L-1-norm. The proposed L-1-MvDA-VC not only can obtain discriminant common space, but also is robust to outliers. In addition, a simple and effective learning algorithm is designed for L-1-MvDA-VC, and its convergence is proved theoretically. Experimental results on real data sets demonstrate that L-1-MvDA-VC has better performance on contaminated data than that of MvDA, MvDA-VC, MvCCDA, and MULDA. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：In this paper, the exponential synchronization of multilayer neural networks (MNNs) is investigated via alternate periodic event-triggered control (APETC). Distinguished from the previous work, a novel APETC which incorporates aperiodically intermittent control (AIC) and periodic event-triggered mechanism is firstly proposed. Determined by two event-triggered conditions, the control and rest intervals of APETC are based on the present state of the system rather than being predetermined. In contrast to the conventional event-triggered control (ETC), the event-triggered conditions of APETC can not only judge the updates of control signals, but also dominate the actuation and close of the controller. Moreover, by introducing the sampling period into ETC, the number of event triggers can be decreased and the Zeno phenomenon is completely avoided. Subsequently, synchro-nization criteria for MNNs under APETC are established on the basis of Lyapunov method and graph theory. Finally, several numerical simulations are performed to demonstrate the validity of the theoretical results. (c) 2022 Published by Elsevier Inc.
NSTL
Elsevier