查看更多>>摘要：Most of the existing feature selection methods tend to pursue the learning performance of the selected feature subset while ignoring the costs of acquiring each feature. However, in real-world problems, we often face the tradeoff between model performance and feature costs because of limited resources. Moreover, in some applications (e.g. medical tests), features are acquired sequentially in the learning process instead of having known the information of the whole feature set in advance. To solve these problems we design a reinforcement learning agent to guide the cost-sensitive feature acquisition process and propose a deep learning-based model to select the informative and lower-cost features for each instance adaptively. The whole process of feature acquisition will be determined by an agent according to what it has observed from inputs. In particular, a Recurrent Neural Network (RNN) model will learn the knowledge from the current sample and the agent will give the instructions on whether the RNN model will continue to select the next feature or stop the sequential feature acquisition process. Moreover, the proposed method can also select the features per block thus it can deal with high dimensional data. We evaluate the effectiveness of the proposed method on a variety of datasets including benchmark datasets, gene datasets, and medical datasets. Compared with the state-of-the-art feature selection methods, the proposed method can achieve comparable learning accuracy while maintaining lower feature costs. (C)& nbsp;2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：The increasing number of people with impairments and the lack of specialists has led to a loss of efficiency to deliver proper treatments from National healthcare systems. In this light, telerehabilitation can play an important role as patients can perform certain therapies at home. Consequently, telerehabilitation systems must support delivering bespoke therapies to patients tailored to their deficits and preferences. However, creating bespoke telerehabilitation activities is a complex and time-consuming task because of the great assortment of deficits. To address this problem, we propose in this research work an automatic generation of such telerehabilitation activities aiming to both assist the specialist in designing and creating telerehabilitation activities that best fit each patient's needs. Therefore, the main contributions of this paper are: (1) the exploitation of Feature Models (FM) to describe the variability in the telerehabilitation domain and to facilitate the communication among the stakeholders to accurately specify the patients' deficits and the features of an association telerehabilitation activity. (2) The design and development of a genetic algorithm (GA) relying on the specified FM able to generate customized association telerehabilitation activities. The FM specified describes precisely the search problem so that the GA chromosomes can be easily identified. It also facilitates the discussion with the stakeholders during the design of the algorithm since its specification can be understood by non-experts in Computer Science. (3) The evaluation of the effectiveness and efficiency of the GA developed by using two sets of experiments: one for tuning the parameters of the GA and another to assess the effectiveness and efficiency of the algorithm while stressed under constraining conditions. (4) The integration of the proposal with a tool for telerehabilitation of people with Acquired Brain Injury (ABI). The proposal targets people with ABI because of the wide assortment of deficits they present, as well as the high impact ABI is having on society, being currently more common than breast cancer, spinal cord injury, HIV/AIDS and multiple sclerosis (MS) combined. (C) 2022 The Author(s). Published by Elsevier B.V.
查看更多>>摘要：In reality, some kinds of time series have the characteristics of high volatility, where fluctuation patterns of sequential data contain rich semantic knowledge and represent the spatial features from two-dimensional perspective. Especially, some non-trivial fluctuations may provide key information for the forecasting of time series. However, how to appropriately represent the fluctuation patterns and achieve the causal reasoning among them remains open. In this work, by learning the causal knowledge of fluctuation patterns, a novel spatial attention fuzzy cognitive map with high-order structure is proposed for the interpretable prediction of time series with high volatility. Firstly, a kind of extended polar fuzzy information granules is utilized to convert time series into granule sequences with interpretable fluctuation features, based on which fuzzy cognitive maps can be constructed by using full data-driven way. Secondly, in order to capture the key fluctuation patterns, the attention mechanism is first introduced into fuzzy cognitive map, where the spatial features of the focused patterns can be taken fully utilized. Thirdly, the high-order structure is involved into the proposed model for the learning of the temporal knowledge existing in the pattern sequences. Finally, real-world financial time series with strong noises and high volatility are empirically utilized to verify the promising performance of the proposed method. (C)& nbsp;2022 Elsevier B.V. All rights reserved.
Bobyr, Maxim, VMilostnaya, Natalia A.Bulatnikov, Valentin A.
13页
查看更多>>摘要：The paper describes a new fuzzy filter based on the method of areas' ratio which allows to reduce noise during filtering signals. To expand the functionality of the method of areas' ratio two computational procedures were developed to eliminate errors inherent in classical defuzzification models, namely a narrow range of defuzzification and insensitivity of a defuzzification model. Presented various computational procedures for the fuzzy filter can change the properties of the output variable resulting. As an example, the proposed mathematical model of the Fuzzy Filter based on the Method of Areas' Ratio illustrated its distinctive characteristics are shown. Firstly, the Fuzzy Filter based on the Method of Areas' Ratio model has the property of continuity. Secondly, computational procedures provide an increase in the performance of the fuzzy filter. Using detailed numerically calculated Root Mean Square Error and Mean Absolute Percentage Error evaluated the proposed model of the fuzzy filter with other filters such as Kalman Filter, Fuzzy Kalman Filter, Ensemble Kalman Filter and Fuzzy Extended Kalman Filter, Basic defuzzification distributions, Fuzzy mean, Quality method, Root mean square and New weighted trapezoid median average. One of the main goals of the article was to confirm the hypothesis about the possibility of using a fuzzy filter based on the method of area's ratio for filtering signals. As well as studies of the sensitivity of the proposed fuzzy filter is based on the Root Mean Square Error and Mean Absolute Percentage Error coefficients. These coefficients were established during the experimental studies and showed that the sensitivity of the fuzzy filter based on the method of area's ratio is better than other filters. (C)& nbsp;2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：Convolutional neural networks (CNNs) and variational models for pansharpening have obtained a compelling performance gain over the state of the art. Inspired by these models, we propose MSCGAN, a generative adversarial network (GAN) with joint multistream architecture and spectral compensation for pansharpening that uses a variational model to incorporate domain-specific knowledge and in particular, introduces a spectral compensation block. First, we extract the structural information of the panchromatic (PAN) image and input it into the generator together with the upsampled multispectral (MS) image. Then, we design a multistream pansharpening CNN architecture suite for domain-specific knowledge. Second, to boost the quality of the pansharpened images, we put the MS image in the generator and design a spectral compensation block. Then, we introduce the concept of the energy function of the variational model and add corresponding spectral constraints and spatial structure constraints to the objective function to achieve a compromise between spectral information fidelity and spatial information fidelity. Finally, the discriminator also introduces spatial structure information to help the generator generate the desired high-resolution multispectral image. Experiments on the images captured by the Quickbird and WorldView-2 satellites show that the our proposed MSCGAN can make use of PAN and LRMS images adequately to attain very encouraging results, obtaining large gains over the state of the art both visually and as measured by quality metrics. (c) 2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：Emergency Landing is a complex problem of optimal path planning of an impaired airplane in presence of obstacles, while the airplane performance characteristics have degraded. Some in-flight failures can affect the airplane dynamics and therefore the new dynamic constraints must be considered in flight planning to the desired landing site. This paper introduces a novel hybrid form of Dubins-simulated annealing (HDSA) optimization framework for emergency landing. The proposed architecture applies Dubins paths and Apollonius' tangent line to generate candidate pieces of trajectories respecting the post-failure performance characteristics of the distressed airplane. The optimization pattern is used to select the optimal combination of the candidate trajectories based on the cost functions and the environmental constraints to lead the airplane to the desired landing site. Analytical performance based equations are developed to achieve an admissible solution in emergency trajectory planning. The goal is to provide a general optimal framework, which can enhance the flight management system by assisting the pilot to plan the most suitable and admissible trajectory to the landing site in emergency flight conditions. The effectiveness of the proposed approach is demonstrated through simulations. (C)& nbsp;2022 Elsevier B.V. All rights reserved.
查看更多>>摘要：Crowd counting using deep convolutional neural networks (CNN) has achieved encouraging progress in recent years. Nevertheless, how to efficiently address the problems of scale variation and complex backgrounds remain a major challenge. For this, we present an innovative Multi-scale Attention Recalibration Network termed MARNet for obtaining more accurate crowd counting. This is achieved mainly by introducing and integrating two significant modules into the proposed model. More concretely, a Feature Pyramid Module (FPM) is first designed to achieve multi-scale feature enhancement by utilizing multiple dilated convolutions with different rates, thus providing rich contextual information for subsequent operations. Besides, to adequately take advantage of these contextual information, a Feature Recalibration Module (FRM) is devised by integrating a Dimension Attention (DA) block with a Region Recalibration (RR) block. The DA block is mainly used for modeling the semantic dependencies between different dimensions of contextual information, while the RR block is responsible for reassigning attention weights for different regions based on the semantic dependencies. By the integration of the above two blocks, the proposed method can be targeted to capture the crowd features for accurately estimating crowd density. Extensive experiments on multiple publicly crowd counting datasets well demonstrate that our method significantly outperforms most existing methods in terms of the counting accuracy and the quality of the generated density map. (c) 2022 Elsevier B.V. All rights reserved.
NSTL
Elsevier