查看更多>>摘要：With the rapid growth of the Internet of Things (IoT), smart systems and applications are equipped with an increasing number of wearable sensors and mobile devices. These sensors are used not only to collect data but, more importantly, to assist in tracking and analyzing the daily human activities. Sensor-based human activity recognition is a hotspot and starts to employ deep learning approaches to supersede traditional shallow learning that rely on hand-crafted features. Although many successful methods have been proposed, there are three challenges to overcome: (1) deep model's performance overly depends on the data size; (2) deep model cannot explicitly capture abundant sample distribution characteristics; (3) deep model cannot jointly consider sample features, sample distribution characteristics, and the relationship between the two. To address these issues, we propose a meta-learning-based graph prototypical model with priority attention mechanism for sensor-based human activity recognition. This approach learns not only sample features and sample distribution characteristics via meta-learning-based graph prototypical model, but also the embeddings derived from priority attention mechanism that mines and utilizes relations between sample features and sample distribution characteristics. What is more, the knowledge learned through our approach can be seen as a priori applicable to improve the performance for other general reasoning tasks. Experimental results on fourteen datasets demonstrate that the proposed approach significantly outperforms other state-of-the-art methods. On the other hand, experiments of applying our model to two other tasks show that our model effectively supports other recognition tasks related to human activity and improves performance on the datasets of these tasks.
查看更多>>摘要：Object detection and location from remote sensing (RS) images is challenging, computationally expensive, and labor intense. Benefiting from research on convolutional neural networks (CNNs), the performance in this field has improved in the recent years. However, object detection methods based on CNNs require a large number of images with annotation information for training. For object location, these annotations must contain bounding boxes. Furthermore, objects in RS images are usually small and densely co-located, leading to a high cost of manual annotation. We tackle the problem of weakly supervised object detection under such conditions, aiming to learn detectors with only image-level annotations, i.e., without bounding box annotations. Based on the fact that the feature maps of a CNN are localizable, we hierarchically fuse the location information from the shallow feature map with the class activation map to obtain accurate object locations. In order to mitigate the loss of small or densely distributed objects, we introduce a divergent activation module and a similarity module into the network. The divergent activation module is used to improve the response strength of the low-response areas in the shallow feature map. Densely distributed objects in RS images, such as aircraft in an airport, often exhibit a certain similarity. The similarity module is used to improve the feature distribution of the shallow feature map and to suppress background noise. Comprehensive experiments on a public dataset and a self-assembled dataset (which we made publicly available) show the superior performance of our method compared to state-of-the-art object detectors.
查看更多>>摘要：When a decision maker is asked to compare a set of alternatives, it may happen that the information provided is incomplete because she has no time to compare all the options or is unable to compare some alternatives against others. This contribution departs from an incomplete fuzzy weak preference relation by completing it on a consistent way with the known information. Herein the original notion of fuzzy transitivity, min-transitivity, is considered. If the decision maker is assumed to be coherent, i.e. under the assumption that the known preferences satisfy transitivity, a complete transitive preference relation that preserves the information given by the coherent decision maker is derived. In the case of the given preference values violate transitivity, a degree of transitivity is defined, and an algorithm is presented to provide the most coherent preference relation that preserves the original information according to that degree of transitivity.
查看更多>>摘要：With the acceleration of the pace of work and life, people are facing more and more pressure, which increases the probability of suffering from depression. However, many patients may fail to get a timely diagnosis due to the serious imbalance in the doctor-patient ratio in the world. A promising development is that physiological and psychological studies have found some differences in speech and facial expression between patients with depression and healthy individuals. Consequently, to improve current medical care, Deep Learning (DL) has been used to extract a representation of depression cues from audio and video for automatic depression detection. To classify and summarize such research, we introduce the databases and describe objective markers for automatic depression estimation. We also review the DL methods for automatic detection of depression to extract a representation of depression from audio and video. Lastly, we discuss challenges and promising directions related to the automatic diagnoses of depression using DL.
查看更多>>摘要：Information fusion refers to derive an overall precise description of data by using certain fusion technique for utilizing the complementary information from multiple sources of data, which can facilitate effective decision-making, prediction and classification, etc. Multi-source homogeneous data, characterizing the data type of variables in the sample in form of one type (i.e., numerical or categorical) in different information sources, which widely exists in many practical applications. This paper concentrates on efficient fusion of multi-source homogeneous data with a data-level fusion model which involves the consolidation of multiple information sources and unsupervised attribute selection of the fused data. A unified description and modeling method of a multi-source homogeneous information system is introduced. The neighborhood rough sets model is used to construct the neighborhood granular structure, which uses the idea of granular computing to build methods of uncertainty measures. Given the uncertainty of fusing multiple information sources, Sup-Inf fusion functions are developed based on the proposed uncertainty measures, which can fuse the multi-source homogeneous information system into a single-source information system. Finally, an unsupervised attribute selection approach is employed to eliminate redundant attribute of the single-source information system. Theoretical analysis and comprehensive experiments on several datasets demonstrate the feasibility and superiority of our method.
Walmsley, Felipe N.Cavalcanti, George D. C.Sabourin, RobertCruz, Rafael M. O....
17页
查看更多>>摘要：In the literature on classification problems, it is widely discussed how the presence of label noise can bring about severe degradation in performance. Several works have applied Prototype Selection techniques, Ensemble Methods, or both, in an attempt to alleviate this issue. Nevertheless, these methods are not always able to sufficiently counteract the effects of noise. In this work, we investigate the effects of noise on a particular class of Ensemble Methods, that of Dynamic Selection algorithms, and we are especially interested in the behavior of the Fire-DES++ algorithm, a state of the art algorithm which applies the Edited Nearest Neighbors (ENN) algorithm to deal with the effects of noise and imbalance. We propose a method which employs multiple Dynamic Selection sets, based on the Bagging-IH algorithm, which we dub Multiple-Set Dynamic Selection (MSDS), in an attempt to supplant the ENN algorithm on the filtering step. We find that almost all methods based on Dynamic Selection are severely affected by the presence of label noise, with the exception of the K-Nearest Oracles-Union algorithm. We also find that our proposed method can alleviate the issues caused by noise in some scenarios. We have made the code for our method available at https://github.com/fnw/baggingds.
查看更多>>摘要：Large-scale group decision-making problems based on social network analysis and minimum cost consensus models (MCCMs) have recently attracted considerable attention. However, few studies have combined them to form a complete decision-making system. Accordingly, we define the satisfaction index to optimize the classical MCCM by considering the effect of the group on individuals. Similarly, we define the consistency index to optimize the consensus reaching process (CRP). Regarding the evolution of the consensus network, the Louvain algorithm is used to divide the entire group into several subgroups to ensure that each subgroup is independent but has strong cohesion. By constructing the MCCM based on the satisfaction index and the optimized consensus-reaching process, the group opinions in each subgroup are ranked to obtain the final ranking of alternatives. Finally, to verify the validity of CRP and the practical value of the proposed model, we conduct consensus network evolution and decision-making analysis in the case of a negotiation between the government and polluting companies to achieve uniform pollution emissions. Sensitivity analysis is performed to demonstrate the stability of the subgroup weights. Furthermore, a comparative analysis using existing models verifies the effectiveness of the proposed model.
查看更多>>摘要：Dwarfism refers to the phenomenon that children with same gender and age are lower than two standard deviations of normal height in the same living environment. It is of great significance for early diagnosis and early treatment of dwarfism. Dwarfism can be divided into growth hormone deficiency (GHD) and idiopathic short stature (ISS). GHD can be distinguished by growth hormone, while ISS is difficult to distinguish because its hormone features are not obvious. Thus, a computer-aided diagnosis model based on brain image data and clinical features is established for the first time and a dwarfism prediction algorithm is proposed based on multimodal pyradiomics. Firstly, we establish the extraction of pituitary gland based on tensor and binary wavelet model, as the pituitary gland is an important organ that affects the growth hormone. Then, the multidimensional fusion model is established to distinguish dwarfism. In the process of distinguishment, the pyradiomics features and clinical features are extracted to distinguish together. Finally, dwarfism computer-aided diagnosis algorithm based on multimodal pyradiomics is realized.
查看更多>>摘要：In machine learning, the model is not as complicated as possible. Good generalization ability means that the model not only performs well on the training data set, but also can make good prediction on new data. Regularization imposes a penalty on model's complexity or smoothness, allowing for good generalization to unseen data even when training on a finite training set or with an inadequate iteration. Deep learning has developed rapidly in recent years. Then the regularization has a broader definition: regularization is a technology aimed at improving the generalization ability of a model. This paper gave a comprehensive study and a state-of-the-art review of the regularization strategies in machine learning. Then the characteristics and comparisons of regularizations were presented. In addition, it discussed how to choose a regularization for the specific task. For specific tasks, it is necessary for regularization technology to have good mathematical characteristics. Meanwhile, new regularization techniques can be constructed by extending and combining existing regularization techniques. Finally, it concluded current opportunities and challenges of regularization technologies, as well as many open concerns and research trends.
查看更多>>摘要：There has recently been growing interest in utilizing multimodal sensors to achieve robust lane line segmentation. In this paper, we introduce a novel multimodal fusion architecture from an information theory perspective, and demonstrate its practical utility using Light Detection and Ranging (LiDAR) camera fusion networks. In particular, we develop, for the first time, a multimodal fusion network as a joint coding model, where each single node, layer, and pipeline is represented as a channel. The forward propagation is thus equal to the information transmission in the channels. Then, we can qualitatively and quantitatively analyze the effect of different fusion approaches. We argue the optimal fusion architecture is related to the essential capacity and its allocation based on the source and channel. To test this multimodal fusion hypothesis, we progressively determine a series of multimodal models based on the proposed fusion methods and evaluate them on the KITTI and the A2D2 datasets. Our optimal fusion network achieves 85%+ lane line accuracy and 98.7%+ overall. The performance gap among the models will inform continuing future research into development of optimal fusion algorithms for the deep multimodal learning community.
NSTL
Elsevier