Hierarchical Aggregation Research Articles

Electricity Load Profile Characterisation for Industrial Users Based on Normal Cloud Model and iCFSFDP Algorithm

Electricity load profile characterisation of industrial users is fundamentally essential for user-side load management and demand response by extracting electricity consumption patterns, which requires load feature-based similarity measurement and accurate curve clustering. Given this background, a typical load curve identification method based on the normal cloud model and the improved clustering by fast search and find of density peaks (iCFSFDP) algorithm is proposed for electricity load profile characterisation of industrial users. First, a piecewise cloud approximation (PWCA) based load dynamic feature extraction algorithm is proposed to establish the piecewise cloud models of daily load curves based on the normal cloud theory, and the overlapping area between clouds is defined to measure the similarity between two load curves for clustering. Second, the iCFSFDP based load curve clustering algorithm is proposed to improve the clustering accuracy by implementing the hierarchical aggregation process. Considering the dispersion and massiveness of metered data in the electricity consumption of industries, a distributed-centralized identification method that extracts the typical curves of each user and the whole industry in a distributed and centralized way respectively is proposed to improve the computational efficiency and the clustering effectiveness. Finally, case studies on the industrial users in Zhejiang province, China show that the proposed method can measure the piecewise power consumption features among load curves comprehensively and determine load curve clusters corresponding to the cloud distance-based similarities, thus helping identify more accurate typical load curves that characterize different electricity consumption profiles.

Aggregation properties of a therapeutic peptide for rheumatoid arthritis: A spectroscopic and molecular dynamics study

The biological properties of therapeutic peptides, such as their pharmacokinetics and pharmacodynamics, are correlated with their structure and aggregation properties. Herein, we studied the aggregation properties of a therapeutic peptide (CIGB-814), currently in phase 2 clinical trial, for the treatment of rheumatoid arthritis over a wide range of concentrations (µM–mM). We applied spectroscopic techniques (fluorescence, circular dichroism, resonance, and dynamic light scattering), atomic force microscopy, and molecular dynamics simulations to determine the aggregation mechanism of CIGB-814. We found that the hierarchical aggregation of CIGB-814 at micromolar concentrations was initiated by the formation of peptide oligomers. Subsequently, the peptide oligomers trigger the nucleation and growth of peptide nanostructures (cac = 123 µM), ultimately leading to the fibrillization of CIGB-814 (cac’ = 508 µM). These results pave the way for a deeper understanding of the CIGB-814 therapeutic activity and may give important insights on its pharmacokinetics.

Saliency map-guided hierarchical dense feature aggregation framework for breast lesion classification using ultrasound image

Deep learning methods, especially convolutional neural networks, have advanced the breast lesion classification task using breast ultrasound (BUS) images. However, constructing a highly-accurate classification model still remains challenging due to complex pattern, relatively-low contrast and fuzzy boundary existing between lesion regions (i.e., foreground) and the surrounding tissues (i.e., background). Few studies have separated foreground and background for learning domain-specific representations, and then fused them for improving performance of models. In this paper, we propose a saliency map-guided hierarchical dense feature aggregation framework for breast lesion classification using BUS images. Specifically, we first generate saliency maps for foreground and background via super-pixel clustering and multi-scale region grouping. Then, a triple-branch network, including two feature extraction branches and a feature aggregation branch, is constructed to learn and fuse discriminative representations under the guidance of priors provided by saliency maps. In particular, two feature extraction branches take the original image and corresponding saliency map as input for extracting foreground- and background-specific representations. Subsequently, a hierarchical feature aggregation branch receives and fuses the features from different stages of two feature extraction branches, for lesion classification in a task-oriented manner. The proposed model was evaluated on three datasets using 5-fold cross validation, and experimental results have demonstrated that it outperforms several state-of-the-art deep learning methods on breast lesion diagnosis using BUS images.

Fast Computation of Born-Approximated Multistatic Scattering.

An algorithm for the fast computation of multistatic scattering patterns, produced by low-contrast inhomogeneous volumetric objects, which satisfy the Born approximation assumption, is presented. The algorithm relies on hierarchical interpolation and aggregation of optimally sampled phase-compensated partial contributions to the scattered field integral. The Born approximation enables the efficient simultaneous computation for ranges of incident and observation angles, as well as excitation frequencies. This leads to a reduction in the asymptotic computational complexity by several polynomial orders of the problem's acoustical size. The algorithm's error controllability and favorable computation time are demonstrated via representative examples.

Edge supervision and multi-scale cost volume for stereo matching

Recently, methods based on Convolutional Neural Network have achieved huge progress in stereo matching. However, it is still difficult to find accurate matching points in inherently ill-posed regions (e.g., weak texture areas and around object edges), in which the accuracy of disparity estimate can be improved by the corresponding geometric constraints. To tackle this problem, we innovatively generate the depth ground-truth boundary dataset by mining the instance segmentation and semantic segmentation datasets and propose RDNet, which incorporates edge cues into stereo matching. The network learns geometric information through a separate processing branch edge stream, which can process feature information in parallel with the stereo stream. The edge stream removes noise and only focuses on processing the relevant boundary information. Besides, we introduce a multi-scale cost volume in hierarchical cost aggregation to enlarge the receptive fields and capture structural and global representations that can significantly improve the ability of scene understanding and disparity estimation accuracy. Moreover, a disparity refinement network with several dilated convolutions is applied to further improve the accuracy of the final disparity estimation. The proposed method is evaluated on Sceneflow, KITTI 2015 and KITTI 2012 benchmark datasets, and the qualitative and quantitative results demonstrate that the proposed RDNet significantly achieves the state-of-the-art stereo matching performance.

Hierarchical Multiresolution Representation of Streaming Time Series

• Generic time series representation model supporting streaming time series data mining at multiple time resolutions. • New representation based on the proposed model, considering fluctuations and continuity of modeled processes. • The application of the proposed model leads to higher and more stable processing speed at lower RAM consumption. • The proposed representation improves subsequent time series clustering. Real-time monitoring, analysis and operations in large industrial systems require an accurate but compact data model created on the basis of a large number of data sources continuously generating massive amounts of data modeled as streaming time series. This paper proposes a generic time series representation approach for reducing data model size and supporting streaming time series data mining at multiple time resolutions. The proposed Hierarchical Multiresolution Time Series Representation model utilizes a buffer-based approach that combines one-pass stream processing with hierarchical aggregation to achieve high processing speed without excessive hardware requirements. In addition, this paper presents a new representation based on the proposed model, Hierarchical Multiresolution Linear-function-based Piecewise Statistical Approximation. The proposed representation considers fluctuations and continuity of modeled processes in order to preserve fundamental characteristics of time series at reduced dimensionality. The usefulness of the proposed solution was proven in a case study. The case study results for generated data set confirm that the proposed model leads to higher and more stable processing speed at lower RAM consumption comparing to related model, especially when dealing with greater number of time resolutions. The case study results for real UK smart meter data confirm that the proposed representation leads to a reduced amount of information loss and an improvement in subsequent time series clustering comparing to related time series representation. Therefore, this paper's main contribution is multiresolution streaming time series data mining support convenient for application in large industrial systems.

Iteration number-based hierarchical gradient aggregation for distributed deep learning

Iteration number-based hierarchical gradient aggregation for distributed deep learning

Mitigating bias in estimating epidemic severity due to heterogeneity of epidemic onset and data aggregation

Outbreaks of infectious diseases, such as influenza, are a major societal burden. Mitigation policies during an outbreak or pandemic are guided by the analysis of data of ongoing or preceding epidemics. The reproduction number, R0, defined as the expected number of secondary infections arising from a single individual in a population of susceptibles is critical to epidemiology. For typical compartmental models such as the Susceptible-Infected-Recovered (SIR) R0 represents the severity of an epidemic. It is an estimate of the early-stage growth rate of an epidemic and is an important threshold parameter used to gain insights into the spread or decay of an outbreak. Models typically use incidence counts as indicators of cases within a single large population; however, epidemic data are the result of a hierarchical aggregation, where incidence counts from spatially separated monitoring sites (or sub-regions) are pooled and used to infer R0. Is this aggregation approach valid when the epidemic has different dynamics across the regions monitored? We characterize bias in the estimation of R0 from a merged data set when the epidemics of the sub-regions, used in the merger, exhibit delays in onset. We propose a method to mitigate this bias, and study its efficacy on synthetic data as well as real-world influenza and COVID-19 data.

Stress‐constrained concurrent topology optimization of two‐scale hierarchical structures

AbstractThis article develops a concurrent topology optimization approach for designing two‐scale hierarchical structures under stress constraints without specifying the topology of the unit cell as a priori. Compared with traditional stress‐constrained topology optimization, the number of stress constraints involved in concurrent topology optimization is much larger and the sensitivity analysis for stress constraints is computationally more expensive. To address these challenges, a novel hierarchical aggregation strategy is proposed to handle the large number of stress constraints and an adjoint method is developed for efficient sensitivity analysis. Several numerical examples are presented to demonstrate the effectiveness of the proposed method. It is also found that the structures obtained by traditional topology optimization usually perform better than the two‐scale structures obtained by concurrent topology optimization when only concentrated loads are considered, while the latter may exhibit better performance in the case that distributed loads are involved.

Visibility analysis of the urbanistic environmet as a constituent of the urbogeosystems approach

This paper recognizes the conceptual approach to the visual analysis of an urbogeosystem (VA UGS), which is within the Urban Studies domain. This approach stays within Human Geography frameworks and based on visibility estimation in the urbanistic environment (UE). The definition of UE is represented as a 3D-formalized model of an actual city environment (CE). In general, our work is a further development of the methodological urbogeosystem concept once introduced by one of this paper’s authors. This concept has been previously explained as a twofold functional entity, which is a base for delineation and analysis of the CE urbogeosystemic properties. The basic possibility of VA UGS has been explained as a possibility, that follows from the circumstance, according to which UE is a quasi-rasterized model of a continual nature of an actual city space and its key features. Then, this model of city continuality can be contrasted with the discrete nature of an urbogeosystem – a hierarchical aggregate of urban features. Exactly this contraposition (continuality versus discrete content) provides the basics for the visual analysis. This research introduces the Visibility Hemisphere concept (VHC), what supposes those conditions of visibility, which do correspond to ambient optic array ones. The Line of Sight (LoS) has been represented as a significant VHC component. Transformation of 2D LoS into 3D-segment of Visibility Hemisphere has been briefly explained within the stereometric frameworks. A number of formalized parameters have been introduced on the base of VHC. Computation of these parameters is the first stage of VA UGS introduction. The introduced approach has been further implemented as the VA-functionality of a web-GIS-application and illustrated with a few interface samples. Visual analysis of Munster-city UGS (Germany) has been provided within this research as a use-case of three scenarios comparative estimation of urban environment developed changes. Possible negative visual impact has been evaluated for each of the scenarios.

Early neoplasia identification in Barrett's esophagus via attentive hierarchical aggregation and self-distillation.

Automatic surveillance of early neoplasia in Barrett's esophagus (BE) is of great significance for improving the survival rate of esophageal cancer. It remains, however, a challenging task due to (1) the large variation of early neoplasia, (2) the existence of hard mimics, (3) the complicated anatomical and lighting environment in endoscopic images, and (4) the intrinsic real-time requirement of this application. We propose a novel end-to-end network equipped with an attentive hierarchical aggregation module and a self-distillation mechanism to comprehensively address these challenges. The hierarchical aggregation module is proposed to capture the complementariness of adjacent layers and hence strengthen the representation capability of each aggregated feature. Meanwhile, an attention mask is developed to selectively integrate the logits of each feature, which not only improves the prediction accuracy but also enhances the prediction interpretability. Furthermore, an efficient self-distillation mechanism is implemented based on a teacher-student architecture, where the student aims at capturing abstract high-level features while the teacher is applied to bring more low-level semantic details to calibrate the classification results. The proposed techniques are effective yet lightweight, improving the classification performance without sacrificing time performance, and thus achieving real-time inference. We extensively evaluate the proposed method on the MICCAI EndoVis Challenge Dataset. Experimental results demonstrate the proposed method can achieve competitive accuracy with a much faster speed than state-of-the-arts.

Hi-EADN: Hierarchical Excitation Aggregation and Disentanglement Frameworks for Action Recognition Based on Videos

Most existing video action recognition methods mainly rely on high-level semantic information from convolutional neural networks (CNNs) but ignore the discrepancies of different information streams. However, it does not normally consider both long-distance aggregations and short-range motions. Thus, to solve these problems, we propose hierarchical excitation aggregation and disentanglement networks (Hi-EADNs), which include multiple frame excitation aggregation (MFEA) and a feature squeeze-and-excitation hierarchical disentanglement (SEHD) module. MFEA specifically uses long-short range motion modelling and calculates the feature-level temporal difference. The SEHD module utilizes these differences to optimize the weights of each spatiotemporal feature and excite motion-sensitive channels. Moreover, without introducing additional parameters, this feature information is processed with a series of squeezes and excitations, and multiple temporal aggregations with neighbourhoods can enhance the interaction of different motion frames. Extensive experimental results confirm our proposed Hi-EADN method effectiveness on the UCF101 and HMDB51 benchmark datasets, where the top-5 accuracy is 93.5% and 76.96%.

Multicomponent Self‐Assembly of a Giant Heterometallic Polyoxotungstate Supercluster with Antitumor Activity

AbstractThe hierarchical aggregation of molecular nanostructures from multiple components is a grand synthetic challenge, which requires highly selective linkage control. We demonstrate how two orthogonal linkage groups, that is, organotin and lanthanide cations, can be used to drive the aggregation of a giant molecular metal oxide superstructure. The title compound {[(Sn(CH3)2)2O]4{[CeW5O18] [TeW4O16][CeSn(CH3)2]4[TeW8O31]4}2}46− (1 a) features dimensions of ca. 2.2×2.3×3.4 nm3 and a molecular weight of ca. 25 kDa. Structural analysis shows the hierarchical aggregation from several independent subunits. Initial biomedical tests show that 1 features an inhibitory effect on the proliferation of HeLa cells based on an apoptosis pathway. In vivo experiments in mice reveal the antiproliferative activity of 1 and open new paths for further development of this new compound class.

Multicomponent Self-Assembly of a Giant Heterometallic Polyoxotungstate Supercluster with Antitumor Activity.

The hierarchical aggregation of molecular nanostructures from multiple components is a grand synthetic challenge, which requires highly selective linkage control. We demonstrate how two orthogonal linkage groups, that is, organotin and lanthanide cations, can be used to drive the aggregation of a giant molecular metal oxide superstructure. The title compound {[(Sn(CH3)2)2O]4{[CeW5O18] [TeW4O16][CeSn(CH3)2]4[TeW8O31]4}2}46− (1 a) features dimensions of ca. 2.2×2.3×3.4 nm3 and a molecular weight of ca. 25 kDa. Structural analysis shows the hierarchical aggregation from several independent subunits. Initial biomedical tests show that 1 features an inhibitory effect on the proliferation of HeLa cells based on an apoptosis pathway. In vivo experiments in mice reveal the antiproliferative activity of 1 and open new paths for further development of this new compound class.

A hierarchical secure data aggregation method using the dragonfly algorithm in wireless sensor networks

A wireless sensor network (WSN) consists of a set of sensor nodes that are widely scattered in inaccessible areas. When deployed in large areas, WSNs generate a large volume of the data. Therefore, efficient methods are needed to process the data. One solution to minimize traffic on large-scale wireless sensor networks is to use data aggregation schemes. In this paper, a secure data aggregation method is proposed. The proposed secure data aggregation scheme has three phases: intra-cluster data aggregation, inter-cluster data aggregation, and data transfer. In the intra-cluster data aggregation phase, a fuzzy scheduling system is designed to adjust the appropriate data transmission rates of the cluster member nodes. In the inter-cluster data aggregation phase, an aggregation tree is created between the cluster head nodes. The dragonfly algorithm (DA) is used to find the optimal aggregation tree between cluster head nodes. In the data transfer phase, the columnar transposition cipher method is used to establish a secure connection between cluster member nodes and their cluster head node. Also, a symmetric and lightweight encryption method based on the residue number system (RNS) is utilized to provide secure communications between the cluster head nodes. We modify RNS and call it RNS+. Finally, the simulation results of the proposed scheme are compared to three data aggregation methods including Sign-share, Sham-share, and RCDA. The results show that the proposed data aggregation scheme outperforms other data aggregation methods in terms of network lifetime, delay and packet delivery rate.

HCANet: A Hierarchical Context Aggregation Network for Semantic Segmentation of High-Resolution Remote Sensing Images

Many practical applications of high-resolution remote sensing images (HRRSIs) are based on semantic segmentation. However, due to the complex ground object information contained in remote sensing images, it is difficult to make precise semantic segmentation of HRRSIs. In this letter, we proposed a hierarchical context aggregation network (HCANet) for the semantic segmentation of HRRSIs. The HCANet has an encoder-decoder structure which is similar to UNet. In the HCANet, we designed two Compact Atrous Spatial Pyramid Pooling (CASPP and CASPP+) modules. The CASPP modules replace the copy and crop operation in UNet to extract the multiscale context information of the multisemantic features of ResNet. The CASPP+ module is embedded in the middle layer of HCANet’s decoder to provide a strong aggregation path of contextual information. In the decoder of HCANet, the multiscale context information obtained by CASPP modules is hierarchically merged layer by layer for the semantic segmentation of HRRSIs. We compared our method with several of the most advanced methods on the ISPRS Vaihingen and Potsdam data sets. The final results demonstrate that our method can achieve outstanding performance.

A Clustering Algorithm via Density Perception and Hierarchical Aggregation Based on Urban Multimodal Big Data for Identifying and Analyzing Categories of Poverty-Stricken Households in China

Nowadays, urban multimodal big data are freely available to the public due to the growing number of cities, which plays a critical role in many fields such as transportation, education, medical treatment, and land resource management. The successful completion of poverty-relief work can greatly improve the quality of people’s life and ensure the sustainable development of the society. Poverty is a severe challenge for human society. It is of great significance to apply machine learning to mine different categories of poverty-stricken households and further provide decision support for poverty alleviation. Traditional poverty alleviation methods need to consume a lot of manpower, material resources, and financial resources. Based on the density-based spatial clustering of applications with noise (DBSCAN), this paper designs the hierarchical DBSCAN clustering algorithm to identify and analyze the categories of poverty-stricken households in China. First, the proposed method adjusts the neighborhood radius dynamically for dividing the data space into several initial clusters with different densities. Then, neighbor clusters are identified by the border and inner distances constantly and aggregated recursively to form new clusters. Based on the idea of division and aggregation, the proposed method can recognize clusters of different forms and deal with noises effectively in the data space with imbalanced density distribution. The experiments indicate that the method has the ideal performance of clustering, which identifies the commonness and difference in characteristics of poverty-stricken households reasonably. In terms of the specific indicator “Accuracy,” the accuracy increases by 2.3% compared with other methods.

Heterogeneous Hierarchical Feature Aggregation Network for Personalized Micro-Video Recommendation

Micro-video recommendation has attracted extensive research attention with the increasing popularity of micro-video sharing platforms. Traditional approaches consider micro-video recommendation as a matching task and ignore the rich relationships among users and micro-videos from various modalities (e.g., visual, acoustic, and textual). Recently, GNN-based approaches show promising performance for the micro-video recommendation task. However, they mainly focus on the homogeneous graph which includes only one type of nodes or relations, and cannot be applied to the heterogeneous graph which consists of users, micro-videos, and related multi-modal information. In this paper, a novel Heterogeneous Hierarchical Feature Aggregation Network (HHFAN) is proposed for personalized micro-video recommendation. Our goal is to explore the highly complicated relationship information among users, micro-videos and related multi-modal information from a modality-aware Heterogeneous Information Graph (M-HIG), and thus generate high-quality user and micro-video embeddings for recommendation. The proposed model consists of two key components: (1) In data structure level, we build a heterogeneous graph and utilize a random walk based sampling strategy to sample neighbors for users and micro-videos. (2) In representation learning level, we design a hierarchical feature aggregation network including the intra- and inter-type feature aggregation networks to better capture the complex structure and rich semantic information in the heterogeneous graph. We evaluate our method on two real-world datasets and the results demonstrate that the proposed model outperforms the baseline methods.

Heat-induced self-assembling of BSA at the isoelectric point

Materials based on ordered protein aggregates have recently received a lot of attention for their application as drug carriers, due to their biocompatibility and their ability to sequester many biological fluids. Bovine serum albumin (BSA) is a good candidate for this use due to its high availability and tendency to aggregate and gel under acidic conditions. In the present work, we employ spectroscopic techniques to investigate the heat-induced BSA aggregation at the molecular scale, in the 12–84 °C temperature range, at pH = 5 where two different isoforms of the protein are stable. Samples at low and high protein concentration are examined. With the advantage of the combined use of FTIR and CD, we recognize the aggregation-prone species and the different distribution of secondary structures, conformational rearrangements and types of aggregates, of millimolar compared to micromolar BSA solutions. Further, as a new tool, we use the Maximum Entropy Method to fit the kinetic curves to investigate the distribution of kinetic constants of the complex hierarchical aggregation process. Finally, we characterize the activation energy of the initial self-assembling step to observe that the formation of both small and large aggregates is driven by the same interactions.

Hierarchical data fusion processes involving the Möbius representation of capacities

The use of the Choquet integral in data fusion processes allows for the effective modelling of interactions and dependencies between data features or criteria. Its application requires identification of the defining capacity (also known as fuzzy measure) values. The main limiting factor is the complexity of the underlying parameter learning problem, which grows exponentially in the number of variables. However, in practice we may have expert knowledge regarding which of the subsets of criteria interact with each other, and which groups are independent. In this paper we study hierarchical aggregation processes, architecturally similar to feed-forward neural networks, but which allow for the simplification of the fitting problem both in terms of the number of variables and monotonicity constraints. We note that the Möbius representation lets us identify a number of relationships between the overall fuzzy measure and the data pipeline structure. Included in our findings are simplified fuzzy measures that generalise both k-intolerant and k-interactive capacities.