Nguyen, Cuong, VLe, Khiem H.Tran, Anh M.Pham, Quang H....
14页
查看更多>>摘要：Transfer learning plays an essential role in Deep Learning, which can remarkably improve the performance of the target domain, whose training data is not sufficient. Our work explores beyond the common practice of transfer learning with a single pre-trained model. We focus on the task of Vietnamese sentiment classification and propose LIFA, a framework to learn a unified embedding from several pre-trained models. We further propose two more LIFA variants that encourage the pre-trained models to either cooperate or compete with one another. Studying these variants sheds light on the success of LIFA by showing that sharing knowledge among the models is more beneficial for transfer learning. Moreover, we construct the AISIA-VN-Review-F dataset, the first large-scale Vietnamese sentiment classification database. We conduct extensive experiments on the AISIA-VN-Review-F and existing benchmarks to demonstrate the efficacy of LIFA compared to other techniques. To contribute to the Vietnamese NLP research, we publish our source code and datasets to the research community upon acceptance. (C) 2021 Elsevier Inc. All rights reserved.
查看更多>>摘要：This paper investigates the problem of designing the optimal completely stealthy attacks in cyber-physical systems. Different from the strictly stealthy attacks in the existing results which still have the possibilities to trigger the alarm and be invalid, a completely stealthy attack model is proposed such that the attack signals are able to bypass the detector successfully without being detected. Under the framework of the attacks, the remote estimation error is analyzed by deriving the recursion of the error covariance matrix, based on which the problem of attack design is transformed into a constrained optimization problem. By employing the Lagrange multiplier method, the optimal attack policy is derived which maximizes the remote estimation error and guarantees the complete stealthiness to the detector concurrently. Finally, simulation examples are provided to illustrate the effectiveness of this work. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：The class of center-based clustering algorithms offers methods to efficiently identify clusters in data sets, making them applicable to larger data sets. While a data set may contain several features, not all of them may be equally informative or helpful towards cluster detection. Therefore, sparse center-based clustering methods offer a way to select only those features that may be useful in identifying the clusters present in a data set. However, to automatically determine the degree to which features should be selected, these methods use the Permutation Method which involves generating and clustering multiple randomly permuted data sets, leading to much higher computation costs. In this paper, we propose an improved approach towards model selection for sparse clustering by using expressions of Bayesian Information Criterion (BIC) derived for the center-based clustering methods of k-Means and Fuzzy c-Means. The derived expressions of BIC require significantly lower computation costs, yet allow us to compare and select a suitable sparse clustering among several possible sparse partitions that may have selected different subsets of features. Experiments on synthetic and real-world data sets show that using BIC for model selection leads to remarkable improvements in the identification of sparse clusterings for both Sparse k-Means and Sparse Fuzzy c-Means. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：In this work, for smoothly achieving the leader-follower consensus formation of multiply omni-mecanum-wheeled robots (OMWRs) with switching desired relative configurations, a self-triggered-organized formation (STOF) system based on distributed model predictive control (DMPC) consensus protocol is proposed. In the STOF system, OMWRs' communication topologies can be timely updated to fit their current formation states through designing the STO condition and updating rules. Through using the self-smoothing transformation mechanism, the jerking of states and velocities in the STOF system can be avoided. After constructing the consensus tracking error system based on the dynamic model of OMWRs, a distributed model predictive control (DMPC) based consensus protocol is utilized to get the optimal input signals for each robot, and the constraints can be synchronically handled. Finally, the results of simulations verify the effectiveness of the developed control scheme. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：Multi-label learning deals with the problem that each instance is associated with multiple labels simultaneously. Most of the existing approaches aim to improve the performance of multi-label learning by exploiting label correlations. Although the data augmentation technique is widely used in many machine learning tasks, it is still unclear whether data augmentation is helpful to multi-label learning. In this article, we propose to leverage the data augmentation technique to improve the performance of multi-label learning. Specifically, we first propose a novel data augmentation approach that performs clustering on the real examples and treats the cluster centers as virtual examples, and these virtual examples naturally embody the local label correlations and label importances. Then, motivated by the cluster assumption that examples in the same cluster should have the same label, we propose a novel regularization term to bridge the gap between the real examples and virtual examples, which can promote the local smoothness of the learning function. Extensive experimental results on a number of real-world multi-label datasets clearly demonstrate that our proposed approach outperforms the state-of-the-art counterparts. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：This paper examines the dependence between electricity prices, demand, and renewable energy sources by means of a multivariate copula model while studying Germany, the widest studied market in Europe. The inter-dependencies are investigated in-depth and monitored over time, with particular emphasis on the tail behavior. To this end, suitable tail dependence measures are introduced to take into account a multivariate extreme scenario appropriately identified through the Kendall's distribution function. The empirical evidence demonstrates a strong association between electricity prices, renewable energy sources, and demand within a day and over the studied years. Hence, this analysis provides guidance for further and different incentives for promoting green energy generation while considering the time-varying dependencies of the involved variables. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：Resampling is the most commonly used method for dealing with imbalanced data, in addition to modifying the algorithm mechanism, it can, for example, generate new minority samples or reduce majority samples to adjust the data distribution. However, to date, related research has predominantly focused on solving the classification problem, while the issue of imbalanced regression data has rarely been discussed. In real-world applications, predicting regression data is a common and valuable issue in decision making, especially in regard to those rare samples with extremely high or low values, such as those encountered in the fields of signal processing, finance, or meteorology. This study therefore divided its regression data into rare samples and normal samples, with self-defined relevance functions and, in addition, proposed a boosting resampling method based on a conditional variational autoencoder. The experimental results showed that when using the proposed resampling method was employed, the prediction performance of the whole testing data set was slightly increased, while the performance for the rare samples was significantly improved. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：In this paper, we propose a new Semi-Nonnegative Matrix Factorization method for 2-dimensional (2D) data, named TS-NMF. It overcomes the drawback of existing methods that seriously damage the spatial information of the data by converting 2D data to vectors in a preprocessing step. In particular, projection matrices are sought under the guidance of building new data representations, such that the spatial information is retained and projections are enhanced by the goal of clustering, which helps construct optimal projection directions. Moreover, to exploit nonlinear structures of the data, manifold is constructed in the projected subspace, which is adaptively updated according to the projections and less afflicted with noise and outliers of the data and thus more representative in the projected space. Hence, seeking projections, building new data representations, and learning manifold are seamlessly integrated in a single model, which mutually enhance other and lead to a powerful data representation. Comprehensive experimental results verify the effectiveness of TS-NMF in comparison with several state-of-the-art algorithms, which suggests high potential of the proposed method for real world applications. (C) 2022 Elsevier Inc. All rights reserved.
查看更多>>摘要：The problem of fault-tolerant controller design for positive Markov jump systems subject to interval uncertainties and time-varying actuator faults is investigated in this paper. First, under the assumption of stochastic stability, an interval observer is developed to achieve simultaneously estimation of the upper and lower bounds of the system state and the actuator faults. Then, an interval observer and a state-estimate stabilizing feedback controller are jointly designed. By utilizing the obtained interval estimates, the controller is constructed to achieve a satisfactory performance and fault tolerance. Owing to positivity, the joint observer and controller design is formulated in terms of linear programming. The effectiveness of the proposed approach is demonstrated through simulation. (C) 2022 Elsevier Inc. All rights reserved.
Park, JinukPark, ChanheeChoi, JonghwanPark, Sanghyun...
21页
查看更多>>摘要：In multivariate time series, a substantial amount of variables exhibit common dynamics stemming from a small number of global factors. Recent studies have shown that the shared information from global components can enhance the forecasting performance of time series. However, existing global-local approaches treat the global factors as additional hidden states inside the model without providing global series for downstream analysis. In this study, we propose DeepGate, a novel time series forecasting framework based on the explicit global-local decomposition. To retain the global and local series property, we have built decomposition and prediction modules separately. In this way, DeepGate can produce interpretable global series for further tasks while improving forecasting performance with the aid of global and local series. In addition, to alleviate the discrepancy between the training and testing steps, we employ a denoising training technique for multi-step forecasting problems. In numerous experiments on real-world benchmarks for time series forecasting, DeepGate outperforms the baselines including existing global-local models. In particular, the experimental results on synthetic tasks demonstrate that our model can effectively extract underlying global series. (C) 2022 Elsevier Inc. All rights reserved.
NSTL
Elsevier