Hierarchical Mechanism Research Articles

Hierarchical multimodal self-attention-based graph neural network for DTI prediction.

Drug-target interactions (DTIs) are a key part of drug development process and their accurate and efficient prediction can significantly boost development efficiency and reduce development time. Recent years have witnessed the rapid advancement of deep learning, resulting in an abundance of deep learning-based models for DTI prediction. However, most of these models used a single representation of drugs and proteins, making it difficult to comprehensively represent their characteristics. Multimodal data fusion can effectively compensate for the limitations of single-modal data. However, existing multimodal models for DTI prediction do not take into account both intra- and inter-modal interactions simultaneously, resulting in limited presentation capabilities of fused features and a reduction in DTI prediction accuracy. A hierarchical multimodal self-attention-based graph neural network for DTI prediction, called HMSA-DTI, is proposed to address multimodal feature fusion. Our proposed HMSA-DTI takes drug SMILES, drug molecular graphs, protein sequences and protein 2-mer sequences as inputs, and utilizes a hierarchical multimodal self-attention mechanism to achieve deep fusion of multimodal features of drugs and proteins, enabling the capture of intra- and inter-modal interactions between drugs and proteins. It is demonstrated that our proposed HMSA-DTI has significant advantages over other baseline methods on multiple evaluation metrics across five benchmark datasets.

Spatiotemporal brain hierarchies of auditory memory recognition and predictive coding

Our brain is constantly extracting, predicting, and recognising key spatiotemporal features of the physical world in order to survive. While neural processing of visuospatial patterns has been extensively studied, the hierarchical brain mechanisms underlying conscious recognition of auditory sequences and the associated prediction errors remain elusive. Using magnetoencephalography (MEG), we describe the brain functioning of 83 participants during recognition of previously memorised musical sequences and systematic variations. The results show feedforward connections originating from auditory cortices, and extending to the hippocampus, anterior cingulate gyrus, and medial cingulate gyrus. Simultaneously, we observe backward connections operating in the opposite direction. Throughout the sequences, the hippocampus and cingulate gyrus maintain the same hierarchical level, except for the final tone, where the cingulate gyrus assumes the top position within the hierarchy. The evoked responses of memorised sequences and variations engage the same hierarchical brain network but systematically differ in terms of temporal dynamics, strength, and polarity. Furthermore, induced-response analysis shows that alpha and beta power is stronger for the variations, while gamma power is enhanced for the memorised sequences. This study expands on the predictive coding theory by providing quantitative evidence of hierarchical brain mechanisms during conscious memory and predictive processing of auditory sequences.

Neural network approaches for rumor stance detection: Simulating complex rumor propagation systems

SummaryThis research introduces a comprehensive suite of neural network models designed to tackle the challenging task of rumor stance detection within the framework of simulating complex rumor propagation systems. Our objective centers on accurately modeling the intricate structures of rumor dialogues and propagation patterns to identify user stances—whether they are in support, denial, questioning, or commenting on rumors. Unlike conventional methods that rely on simplistic keyword targeting and fail in the nuanced context of social networks, our models delve into the complexities of dialogue and propagation structures, offering a more precise and insightful analysis of rumor dynamics. In addressing the simulation and modeling of complex systems, our approach specifically focuses on the elaborate interaction networks that underpin rumor spread and reception. While our methodology does not directly engage with brain‐like computing paradigms, it reflects a similar level of sophistication in handling layered and complex information flows, analogous to cognitive processes in understanding and interpreting human communications. Employing a hierarchical attention mechanism, our models adeptly parse through multitiered dialogue sequences, effectively distinguishing between various indicators of user stances. This allows for a nuanced and detailed representation of the rumor ecosystem, significantly enhancing the accuracy of stance detection. Through rigorous testing on diverse datasets, our approach has demonstrated superior performance over existing models, thereby establishing a new benchmark in the field.

Data-efficient multimodal human action recognition for proactive human–robot collaborative assembly: A cross-domain few-shot learning approach

With the recent vision of Industry 5.0, the cognitive capability of robots plays a crucial role in advancing proactive human–robot collaborative assembly. As a basis of the mutual empathy, the understanding of a human operator’s intention has been primarily studied through the technique of human action recognition. Existing deep learning-based methods demonstrate remarkable efficacy in handling information-rich data such as physiological measurements and videos, where the latter category represents a more natural perception input. However, deploying these methods in new unseen assembly scenarios requires first collecting abundant case-specific data. This leads to significant manual effort and poor flexibility. To deal with the issue, this paper proposes a novel cross-domain few-shot learning method for data-efficient multimodal human action recognition. A hierarchical data fusion mechanism is designed to jointly leverage the skeletons, RGB images and depth maps with complementary information. Then a temporal CrossTransformer is developed to enable the action recognition with very limited amount of data. Lightweight domain adapters are integrated to further improve the generalization with fast finetuning. Extensive experiments on a real car engine assembly case show the superior performance of proposed method over state-of-the-art regarding both accuracy and finetuning efficiency. Real-time demonstrations and ablation study further indicate the potential of early recognition, which is beneficial for the robot procedures generation in practical applications. In summary, this paper contributes to the rarely explored realm of data-efficient human action recognition for proactive human–robot collaboration.

Hierarchical spatial attention-based cross-scale detection network for Digital Works Supervision System (DWSS)

Vision-based on-site detection is a crucial component of the Digital Works Supervision System (DWSS). However, accurate detecting cross-scale objects in complicated construction sites remains a challenge. In this research, a hierarchical spatial attention-based cross-scale detection network is designed to address these challenges and provide accurate on-site information for the DWSS. Firstly, a convolutional neural network is constructed to extract multi-scale feature maps with different levels of information. Furthermore, a hierarchical spatial attention mechanism is proposed to facilitate information propagation and complement between multi-scale feature maps. Secondly, the cascade detection mechanism is designed to improve the detection performance of cross-scale objects. Large-scale feature maps with more detailed information are used for the detection of small-scale objects such as workers. Small-scale feature maps with semantic features are used for the detection of large-scale objects such as construction machines. Finally, the on-site information is automatically extracted from detection results and converted into suitable data formats to generate on-site reports for the DWSS. Experiments demonstrate that the proposed method can achieve SOTA detection performance on both MOCS and SODA dataset, especially for the small-scale objects and complicated construction scenarios.

An intelligent assessment method of criminal psychological attribution based on unbalance data

Criminal cases often exhibit imbalance and cannot be extended by data augmentation when classified into attribution types. To solve the problem of unbalance data in offenders’ attribution classification, this paper proposes a criminal psychological attribution assessment model by an improved Balanced TF-Distinguishing IDF method (B-TF-dIDF) and constructed a hybrid network with attention method to fuse numerical and text features for improving the accuracy. First, as a statistical method, B-TF-dIDF is presented to reduce the impact of class-imbalance for extraction of numerical features, in which a balanced element is added to reduce the effects of incorrect type keywords on classification, and a distinguishing element is added to discriminate the types of keywords. Then, an improved hybrid network model composed of Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) is constructed to balance the influence of different lengths of text samples for extracting the semantic features of criminal texts. For evaluating different feature weights by their importance, Spatial Attention is used to improve CNN in the feature maps. Moreover, the self-attention is also performed to re-evaluate the mixed features. Finally, the softmax classifier provides a scientific basis for developing a hierarchical treatment mechanism further. Additionally, we build a criminal data set with labels from real cases for testing. The experiment proved that the proposed model is better than other related methods in various evaluation indicators, including the micro and macro scopes. Moreover, the F1 of minority samples has increased by 6%–8%, indicating that the proposed method can reduce the impact of class-imbalance.

Two Level Security for Key Access Using Cloud Computing

In this work, the author constructs a key access management scheme that seamlessly transitions any hierarchical-like access policy to the digital medium. The proposed scheme allows any public cloud system to be used as a private cloud. We consider the data owner an entity consisting of several organization units. We provide a secure method for each user of this entity to access the public cloud from both inside and outside the company’s network. The idea of the key access control scheme, which is based on Shamir’s secret sharing algorithm and polynomial interpolation method, is suitable especially for hierarchical organizational structures. It offers a secure, flexible, and hierarchical key access mechanism for organizations utilizing mission-critical data.

Query-Adaptive Late Fusion for Hierarchical Fine-Grained Video-Text Retrieval.

Recently, a hierarchical fine-grained fusion mechanism has been proved effective in cross-modal retrieval between videos and texts. Generally, the hierarchical fine-grained semantic representations (video-text semantic matching is decomposed into three levels including global-event representation matching, action-relation representation matching, and local-entity representation matching) to be fused can work well by themselves for the query. However, in real-world scenarios and applications, existing methods failed to adaptively estimate the effectiveness of multiple levels of the semantic representations for a given query in advance of multilevel fusion, resulting in a worse performance than expected. As a result, it is extremely essential to identify the effectiveness of hierarchical semantic representations in a query-adaptive manner. To this end, this article proposes an effective query-adaptive multilevel fusion (QAMF) model based on manipulating multiple similarity scores between the hierarchical visual and text representations. First, we decompose video-side and text-side representations into hierarchical semantic representations consisting of global-event level, action-relation level, and local-entity level, respectively. Then, the multilevel representation of the video-text pair is aligned to calculate the similarity score for each level. Meanwhile, the sorted similarity score curves of the good semantic representation are different from the inferior ones, which exhibit a "cliff" shape and gradually decline (see Fig. fig1 as an example). Finally, we leverage the Gaussian decay function to fit the tail of the score curve and calculate the area under the normalized sorted similarity curve as the indicator of semantic representation effectiveness, namely, the area of good semantic representation is small, and vice versa. Extensive experiments on three public benchmark video-text datasets have demonstrated that our method consistently outperforms the state-of-the-art (SoTA). A simple demo of QAMF will soon be publicly available on our homepage: https://github.com/Lab-ANT.

Hierarchical Scheduling Mechanism Based on Relaxed Constraints for Complex Industrial Scenes

Hierarchical Scheduling Mechanism Based on Relaxed Constraints for Complex Industrial Scenes

HCA-YOLO: a non-salient object detection method based on hierarchical attention mechanism

HCA-YOLO: a non-salient object detection method based on hierarchical attention mechanism

Improvement of Practical Byzantine Fault Tolerance Consensus Algorithm Based on DIANA in Intellectual Property Environment Transactions

In response to the shortcomings of the consensus algorithm for intellectual property transactions, such as high communication overhead, random primary node selection, and prolonged consensus time, a Practical Byzantine Fault Tolerance (PBFT) improvement algorithm based on Divisive Analysis (DIANA) D-PBFT algorithm is proposed. Firstly, the algorithm adopts the hierarchical clustering mechanism of DIANA to cluster nodes based on similarity, enhancing node partition accuracy and reducing the number of participating consensus nodes. Secondly, it designs a reward and punishment system based on node ranking, to achieve consistency between node status and permissions, timely evaluation, and feedback on node behaviours, thereby enhancing node enthusiasm. Then, the election method of the primary node is improved by constructing proxy and alternate nodes and adopting a majority voting strategy to achieve the selection and reliability of the primary node. Finally, the consistency protocol is optimised to perform consensus once within the cluster and once between all primary nodes, to ensure the accuracy of the consensus results. Experimental results demonstrate that the D-PBFT algorithm shows a better performance, in terms of communication complexity, throughput, and latency.

Precise cortical contributions to sensorimotor feedback control during reactive balance.

The role of the cortex in shaping automatic whole-body motor behaviors such as walking and balance is poorly understood. Gait and balance are typically mediated through subcortical circuits, with the cortex becoming engaged as needed on an individual basis by task difficulty and complexity. However, we lack a mechanistic understanding of how increased cortical contribution to whole-body movements shapes motor output. Here we use reactive balance recovery as a paradigm to identify relationships between hierarchical control mechanisms and their engagement across balance tasks of increasing difficulty in young adults. We hypothesize that parallel sensorimotor feedback loops engaging subcortical and cortical circuits contribute to balance-correcting muscle activity, and that the involvement of cortical circuits increases with balance challenge. We decomposed balance-correcting muscle activity based on hypothesized subcortically- and cortically-mediated feedback components driven by similar sensory information, but with different loop delays. The initial balance-correcting muscle activity was engaged at all levels of balance difficulty. Its onset latency was consistent with subcortical sensorimotor loops observed in the lower limb. An even later, presumed, cortically-mediated burst of muscle activity became additionally engaged as balance task difficulty increased, at latencies consistent with longer transcortical sensorimotor loops. We further demonstrate that evoked cortical activity in central midline areas measured using electroencephalography (EEG) can be explained by a similar sensory transformation as muscle activity but at a delay consistent with its role in a transcortical loop driving later cortical contributions to balance-correcting muscle activity. These results demonstrate that a neuromechanical model of muscle activity can be used to infer cortical contributions to muscle activity without recording brain activity. Our model may provide a useful framework for evaluating changes in cortical contributions to balance that are associated with falls in older adults and in neurological disorders such as Parkinson's disease.

Hierarchical Cross Traffic Scheduling Based on Time-Aware Shapers for Mobile Time-Sensitive Fronthaul Network

To solve the problem of jitter and low network throughput caused by the impact of background flows on IQ traffic in mobile fronthaul network, this paper proposed a new scheduling model for background flows, named hierarchical crossover traffic scheduling mechanism based on time-aware shaper (HC-TAS) by improving the traditional counterpart. Then, in this new model, we designed an inbound scheduling algorithm based on frame length matching and an outbound scheduling algorithm based on queue status, making sure that smaller data frames will not be blocked by large data frames. This greatly improves the utilization of timeslots in the scheduling process and reduces the jitter impact of background flows. To verify its performance, we conducted experiments in a simulated fronthaul network conforming to IEEE 802.1CM. The experimental results show that, under the condition that the jitter is guaranteed to be zero, compared with two mainstream scheduling schemes, Comb-FITting and TAS + Preemption, our proposed scheme can achieve lower maximum end-to-end delay and higher link utilization. The proposed HC-TAS meets the requirements of low jitter and high bandwidth utilization in 5G fronthaul network, and the research results provide a technical basis for the application and development of general time-sensitive networks as well.

HSAA-CD: A Hierarchical Semantic Aggregation Mechanism and Attention Module for Non-Agricultural Change Detection in Cultivated Land

Cultivated land plays a fundamental role in the sustainable development of the world. Monitoring the non-agricultural changes is important for the development of land-use policies. A bitemporal image transformer (BIT) can achieve high accuracy for change detection (CD) tasks and also become a key scientific tool to support decision-making. Because of the diversity of high-resolution RSIs in series, the complexity of agricultural types, and the irregularity of hierarchical semantics in different types of changes, the accuracy of non-agricultural CD is far below the need for the management of the land and for resource planning. In this paper, we proposed a novel non-agricultural CD method to improve the accuracy of machine processing. First, multi-resource surveying data are collected to produce a well-tagged dataset with cultivated land and non-agricultural changes. Secondly, a hierarchical semantic aggregation mechanism and attention module (HSAA) bitemporal image transformer method named HSAA-CD is performed for non-agricultural CD in cultivated land. The proposed HSAA-CD added a hierarchical semantic aggregation mechanism for clustering the input data for U-Net as the backbone network and an attention module to improve the feature edge. Experiments were performed on the open-source LEVIR-CD and WHU Building-CD datasets as well as on the self-built RSI dataset. The F1-score, intersection over union (IoU), and overall accuracy (OA) of these three datasets were 88.56%, 84.29%, and 68.50%; 79.84%, 73.41%, and 59.29%; and 98.83%, 98.39%, and 93.56%, respectively. The results indicated that the proposed HSAA-CD method outperformed the BIT and some other state-of-the-art methods and proved to be suitable accuracy for non-agricultural CD in cultivated land.

Citation Forecasting with Multi-Context Attention-Aided Dependency Modeling

Forecasting citations of scientific patents and publications is a crucial task for understanding the evolution and development of technological domains and for foresight into emerging technologies. By construing citations as a time series, the task can be cast into the domain of temporal point processes. Most existing work on forecasting with temporal point processes, both conventional and neural network-based, only performs single-step forecasting. In citation forecasting, however, the more salient goal is n -step forecasting: predicting the arrival of the next n citations. In this article, we propose Dynamic Multi-Context Attention Networks (DMA-Nets), a novel deep learning sequence-to-sequence (Seq2Seq) model with a novel hierarchical dynamic attention mechanism for long-term citation forecasting. Extensive experiments on two real-world datasets demonstrate that the proposed model learns better representations of conditional dependencies over historical sequences compared to state-of-the-art counterparts and thus achieves significant performance for citation predictions.

Dynamic interplay of microtubule and actomyosin forces drive tissue extension

In order to shape a tissue, individual cell-based mechanical forces have to be integrated into a global force pattern. Over the last decades, the importance of actomyosin contractile arrays, which are the key constituents of various morphogenetic processes, has been established for many tissues. Recent studies have demonstrated that the microtubule cytoskeleton mediates folding and elongation of the epithelial sheet during Drosophila morphogenesis, placing microtubule mechanics on par with actin-based processes. While these studies establish the importance of both cytoskeletal systems during cell and tissue rearrangements, a mechanistic understanding of their functional hierarchy is currently missing. Here, we dissect the individual roles of these two key generators of mechanical forces during epithelium elongation in the developing Drosophila wing. We show that wing extension, which entails columnar-to-cuboidal cell shape remodeling in a cell-autonomous manner, is driven by anisotropic cell expansion caused by the remodeling of the microtubule cytoskeleton from apico-basal to planarly polarized. Importantly, cell and tissue elongation is not associated with Myosin activity. Instead, Myosin II exhibits a homeostatic role, as actomyosin contraction balances polarized microtubule-based forces to determine the final cell shape. Using a reductionist model, we confirm that pairing microtubule and actomyosin-based forces is sufficient to recapitulate cell elongation and the final cell shape. These results support a hierarchical mechanism whereby microtubule-based forces in some epithelial systems prime actomyosin-generated forces.

Robotic Odor Source Localization via Vision and Olfaction Fusion Navigation Algorithm.

Robotic odor source localization (OSL) is a technology that enables mobile robots or autonomous vehicles to find an odor source in unknown environments. An effective navigation algorithm that guides the robot to approach the odor source is the key to successfully locating the odor source. While traditional OSL approaches primarily utilize an olfaction-only strategy, guiding robots to find the odor source by tracing emitted odor plumes, our work introduces a fusion navigation algorithm that combines both vision and olfaction-based techniques. This hybrid approach addresses challenges such as turbulent airflow, which disrupts olfaction sensing, and physical obstacles inside the search area, which may impede vision detection. In this work, we propose a hierarchical control mechanism that dynamically shifts the robot's search behavior among four strategies: crosswind maneuver, Obstacle-Avoid Navigation, Vision-Based Navigation, and Olfaction-Based Navigation. Our methodology includes a custom-trained deep-learning model for visual target detection and a moth-inspired algorithm for Olfaction-Based Navigation. To assess the effectiveness of our approach, we implemented the proposed algorithm on a mobile robot in a search environment with obstacles. Experimental results demonstrate that our Vision and Olfaction Fusion algorithm significantly outperforms vision-only and olfaction-only methods, reducing average search time by 54% and 30%, respectively.

Energy Consumption Optimization based on Economic Benefit in WSN-based IoT via Global Hierarchical Caching Strategy

Recent advances in Wireless Sensor Network (WSN) have significantly contributed to the prevalence of Internet of Things (IoT) devices and their pervasive integration into everyday life and industrial operations. A WSN-based IoT comprises numerous small, scattered, battery-operated sensors that are designed to perform collaborative tasks. These sensor nodes are prone to energy drain because of their limited battery capacity when they needed to run efficiently over extended periods of time. Enhancing the large-scale network’s lifespan and controlling the economic costs become relevant since battery replacement or recharging is impractical in severe environments. In this paper, we present a novel energy optimization algorithm tailored for WSNs, which not only takes into account the reduction in replacement costs stemming from prolonged equipment lifespan but also incorporates the operational savings resulting from enhanced energy efficiency. It aims to enhance energy efficiency by leveraging a global hierarchical caching mechanism to simultaneously balance exploration and exploitation of energy resources in both the uplink and downlink of the networks. The simulation results demonstrate that our algorithm effectively minimizes energy consumption while maintaining optimal economic efficiency by decreasing the frequency of state transitions. It can consume 13% less energy than original system and extend the network lifetime by 10%.

Hierarchical Patch Aggregation Transformer for Motion Deblurring

The encoder-decoder framework based on Transformer components has become a paradigm in the field of image deblurring architecture design. In this paper, we critically revisit this approach and find that many current architectures severely focus on limited local regions during the feature extraction stage. These designs compromise the feature richness and diversity of the encoder-decoder framework, leading to bottlenecks in performance improvement. To address these deficiencies, a novel Hierarchical Patch Aggregation Transformer architecture (HPAT) is proposed. In the initial feature extraction stage, HPAT combines Axis-Selective Transformer Blocks with linear complexity and is supplemented by an adaptive hierarchical attention fusion mechanism. These mechanisms enable the model to effectively capture the spatial relationships between features and integrate features from different hierarchical levels. Then, we redesign the feedforward network of the Transformer block in the encoder-decoder structure and propose the Fused Feedforward Network. This effective aggregation enhances the ability to capture and retain local detailed features. We evaluate HPAT through extensive experiments and compare its performance with baseline methods on public datasets. Experimental results show that the proposed HPAT model achieves state-of-the-art performance in image deblurring tasks.

Hierarchical and dynamic graph attention network for drug-disease association prediction.

In the realm of biomedicine, the prediction of associations between drugs and diseases holds significant importance. Yet, conventional wet lab experiments often fall short of meeting the stringent demands for prediction accuracy and efficiency. Many prior studies have predominantly focused on drug and disease similarities to predict drug-disease associations, but overlooking the crucial interactions between drugs and diseases that are essential for enhancing prediction accuracy. Hence, in this paper, a resilient and effective model named Hierarchical and Dynamic Graph Attention Network (HDGAT) has been proposed to predict drug-disease associations. Firstly, it establishes a heterogeneous graph by leveraging the interplay of drug and disease similarities and associations. Subsequently, it harnesses the capabilities of graph convolutional networks and bidirectional long short-term memory networks (Bi-LSTM) to aggregate node-level information within the heterogeneous graph comprehensively. Furthermore, it incorporates a hierarchical attention mechanism between convolutional layers and a dynamic attention mechanism between nodes to learn embeddings for drugs and diseases. The hierarchical attention mechanism assigns varying weights to embeddings learned from different convolutional layers, and the dynamic attention mechanism efficiently prioritizes inter-node information by allocating each node with varying rankings of attention coefficients for neighbour nodes. Moreover, it employs residual connections to alleviate the over-smoothing issue in graph convolution operations. The latent drug-disease associations are quantified through the fusion of these embeddings ultimately. By conducting 5-fold cross-validation, HDGAT's performance surpasses the performance of existing state-of-the-art models across various evaluation metrics, which substantiates the exceptional efficacy of HDGAT in predicting drug-disease associations.