Local Model Research Articles

A Study of Improved Two-Stage Dual-Conv Coordinate Attention Model for Sound Event Detection and Localization.

Sound Event Detection and Localization (SELD) is a comprehensive task that aims to solve the subtasks of Sound Event Detection (SED) and Sound Source Localization (SSL) simultaneously. The task of SELD lies in the need to solve both sound recognition and spatial localization problems, and different categories of sound events may overlap in time and space, making it more difficult for the model to distinguish between different events occurring at the same time and to locate the sound source. In this study, the Dual-conv Coordinate Attention Module (DCAM) combines dual convolutional blocks and Coordinate Attention, and based on this, the network architecture based on the two-stage strategy is improved to form the SELD-oriented Two-Stage Dual-conv Coordinate Attention Model (TDCAM) for SELD. TDCAM draws on the concepts of Visual Geometry Group (VGG) networks and Coordinate Attention to effectively capture critical local information by focusing on the coordinate space information of the feature map and dealing with the relationship between the feature map channels to enhance the feature selection capability of the model. To address the limitation of a single-layer Bi-directional Gated Recurrent Unit (Bi-GRU) in the two-stage network in terms of timing processing, we add to the structure of the two-layer Bi-GRU and introduce the data enhancement techniques of the frequency mask and time mask to improve the modeling and generalization ability of the model for timing features. Through experimental validation on the TAU Spatial Sound Events 2019 development dataset, our approach significantly improves the performance of SELD compared to the two-stage network baseline model. Furthermore, the effectiveness of DCAM and the two-layer Bi-GRU structure is confirmed by performing ablation experiments.

Missile Fault Detection and Localization Based on HBOS and Hierarchical Signed Directed Graph

The rudder surfaces and lifting surfaces of a missile are utilized to acquire aerodynamic forces and moments, adjust the missile’s attitude, and achieve precise strike missions. However, the harsh flying conditions of missiles make the rudder surfaces and lifting surfaces susceptible to faults. In practical scenarios, there is often a scarcity of fault data, and sometimes, it is even difficult to obtain such data. Currently, data-driven fault detection and localization methods heavily rely on fault data, posing challenges for their applicability. To address this issue, this paper proposes an HBOS (Histogram-Based Outlier Score) online fault-detection method based on statistical distribution. This method generates a fault-detection model by fitting the probability distribution of normal data and incorporates an adaptive threshold to achieve real-time fault detection. Furthermore, this paper abstracts the interrelationships between the missile’s flight states and the propagation mechanism of faults into a hierarchical directed graph model. By utilizing bilateral adaptive thresholds, it captures the first fault features of each sub-node and determines the fault propagation effectiveness of each layer node based on the compatibility path principle, thus establishing a fault inference and localization model. The results of semi-physical simulation experiments demonstrate that the proposed algorithm is independent of fault data and exhibits high real-time performance. In multiple sets of simulated tests with randomly parameterized deviations, the fault-detection accuracy exceeds 98% with a false-alarm rate of no more than 0.31%. The fault-localization algorithm achieves an accuracy rate of no less than 97.91%.

3D winding path modeling method with fiber overlap effect

The fibers will repeatedly cross and stack on the liner during the continuous winding process, which poses a challenge in creating accurate local and global structure models. To address the issue, a 3D winding path modeling method is proposed to simulate the fiber overlap effect. First, the initial path is offset equidistantly along the direction of width and thickness to construct the 3D path points. Then, by calculating the overlap thickness, the overlapped fibers are raised to the corresponding overlap thickness to eliminate interference in the 3D path points. Finally, a 3D winding path with the fiber overlap effect is obtained by solving the suspended points before and after the overlapping area to describe the behavior of fibers separating from the liner surface under tension. The proposed method is applied to various winding patterns and liner shapes, and the results show that the established 3D winding path can accurately simulate the fiber overlap effect and correspond with the actual local and global winding structures.

Communication-Efficient Wireless Traffic Prediction with Federated Learning

Wireless traffic prediction is essential to developing intelligent communication networks that facilitate efficient resource allocation. Along this line, decentralized wireless traffic prediction under the paradigm of federated learning is becoming increasingly significant. Compared to traditional centralized learning, federated learning satisfies network operators’ requirements for sensitive data protection and reduces the consumption of network resources. In this paper, we propose a novel communication-efficient federated learning framework, named FedCE, by developing a gradient compression scheme and an adaptive aggregation strategy for wireless traffic prediction. FedCE achieves gradient compression through top-K sparsification and can largely relieve the communication burdens between local clients and the central server, making it communication-efficient. An adaptive aggregation strategy is designed by quantifying the different contributions of local models to the global model, making FedCE aware of spatial dependencies among various local clients. We validate the effectiveness of FedCE on two real-world datasets. The results demonstrate that FedCE can improve prediction accuracy by approximately 27% with only 20% of communications in the baseline method.

Interaction Effects on the Dynamical Anderson Metal-Insulator Transition Using Kicked Quantum Gases.

Understanding the interplay of interaction and disorder in quantum transport poses long-standing scientific challenges for theory and experiment. While highly controlled ultracold atomic platforms combining atomic interactions with spatially disordered lattices have led to remarkable advances, the extension of such controlled studies to phenomena in high-dimensional disordered systems, such as the three-dimensional Anderson metal-insulator transition has been limited. Kicked quantum gases provide an alternate experimental platform that captures the Anderson model in momentum space and features dynamical localization as the analog of Anderson localization. Here, we utilize a momentum space lattice platform using quasiperiodically kicked ultracold atomic gases to experimentally investigate interaction effects on the three-dimensional dynamical Anderson metal-insulator transition. We observe interaction-driven subdiffusion and a divergence of delocalization onset time on approaching the phase boundary. Mean-field numerical simulations show qualitative agreement with experimental observations, but with significant quantitative deviations.

Model Pengembangan Tenaga Kerja Lokal dalam Meningkatkan Perekonomian dan Menunjang Pariwisata Desa Wisata Saba Kecamatan Blahbatuh Kabupaten Gianyar

The legal protection of local workers in the Law of the Republic of Indonesia Number 13 of 2003 concerning Manpower and Regional Regulation (Perda) of Bali Province Number 10 of 2019 concerning the implementation of Manpower, Law Number 10 of 2009 concerning Tourism, Regional Regulation of Bali Province Number 5 of 2020 concerning Standards for Organizing Balinese Cultural Tourism, in companies located in Saba Tourism Village, Blahbatuh District, Gianyar Regency is a study of researchers. Companies located in Saba Tourism Village, Blahbatuh District, Gianyar Regency employ local workers in all fields, and it is highly expected that the Government in general and employers as investors and local workers in particular that arrangements regarding legal protection and local labor development models require separate policies from employers, where the employer is also obliged to foster good relations with all parties, namely investors, the Saba Tourism Village community and the Government of Gianyar Regency, because the policies and regulations issued by the company are inseparable from its supervision, to avoid the possibility of arbitrary actions by employers against the local workforce itself. The method used is empirical legal research, with primary data through field research, secondary data through library research. The purpose of this study is to examine and analyze the local workforce development model in the Company located in Saba tourist attraction Blahbatuh District Gianyar Regency, as well as the constraints and obstacles faced by a company in applying the legislation. From the research results obtained that the local labor development model seen from the condition of the internal factors of local labor development in improving the economy of the Saba tourist village community is still greater than the weakness factors, while the condition of external factors is still greater than the opportunity factor threat. The local workforce development strategy in improving the economy is a wearknes-opportunity (WO) strategy, namely improving the skills of local workers in the formal or informal sector, especially for local workers, then improving facilities and infrastructure and supporting human resources (HR), the employment training center (BLK) for training and improving the business skills of local workers in various sectors, accelerating services and prioritizing business investment rules so that they can absorb more local labor. While the empowerment of local labor in the Krisna souvenir business, Ajik's milk pie business, Aloe Vera business, Villa and Restorant business, Turtle Breeder business, Plaminggo business located in the Saba tourist village, Gianyar Regency, in accordance with the Bali Provincial Regional Regulation Number 10 of 2019 concerning the implementation of Manpower has agreed with the Blangsinga traditional village to coordinate with the Customary Village and it has been agreed to employ local labor around 70% from local labor in the Blangsinga traditional village, and 30% from outside labor. This shows that the local workforce recruited comes from their birthplace or is native to the area where they live and are domiciled in the area and are proven by identity cards (KTP) and family cards (KK). As well as the role of traditional villages in empowering local labor is very large, very influential on the company's commitment and there must be approval from the traditional village every time they recruit workers, with the aim that there is a balance between the rights and obligations of companies operating in the village which contributes to the welfare of the surrounding community and automatically provides welfare from the economic aspect. The targeted scientific journal output is a Sinta 4 Accredited National Journal entitled KertaWicaksana Journal, as well as a 2024 research report, and Intellectual Property Rights (IPR).

Reference universe and competition in higher education: Israeli higher education organizations constituting excellence

ABSTRACT Seeing that the culture and governance of competition determines relations among social actors, we turn to the concept of referentiality to investigate the institutional order of competition in higher education. Defining referentiality as the description of a model standard for comparison, this study explores how higher education organizations (HEOs), specifically universities and colleges in Israel, conceive of their positions in a landscape of multiple competitions. Through text analysis of 33 excellence statements made by Israeli HEOs since 2020, the study identifies components of referentiality, namely the substantive dimension of excellence (such as research, innovation, teaching, or sports), the basis for referencing (ranking scales and agencies), and the referent (global or local models). Findings reveal that the multiplicity of components for comparability, as well as the multiple categories of HEOs, create a complex array of referentiality and competition and outline the reference universe of Israeli HEOs. Our study underscores the performative and agentic nature of referentiality in shaping organizational identity and status in the expansive and highly competitive global academic field.

Training-free approach to constructing ensemble of local experts

Conventional machine learning methods assume that a central server holds the entire training dataset for training a global model. However, in many real-world situations, datasets are collected from individual local nodes in distributed environments. Gathering these datasets often poses challenges owing to various practical constraints such as data security and privacy issues, storage limits, and transmission costs. When data distributions differ among local nodes, local experts trained on individual local nodes exhibit poor generalization performance for new instances. In this study, we propose a training-free method for constructing an ensemble of local experts without accessing datasets. For a query instance, we apply uncertainty quantification and out-of-distribution (OOD) detection to calculate the uncertainty and OOD scores of the predictions from individual models. We then determine the weights of the models by assigning a lower weight to those with higher uncertainty and OOD scores. Using these weights, the predictions of the individual models are aggregated to obtain a final prediction. In contrast to existing methods that impose specific application requirements, the proposed method can be applied in situations in which only individual predictions from local models are available for use when making a prediction for a query instance. We evaluate the effectiveness of the proposed method using image classification benchmark datasets.

A systemic analysis of participatory land and water governance in the Tsitsa River Catchment, South Africa

AbstractEffective governance of complex social‐ecological systems requires diverse stakeholders to be engaged in meaningful participatory processes. Meaningful participation must be intelligible and significant to local actors and coherent within a broader, multilevel governance system. This study explores the complexities of participatory land and water governance in South Africa's Tsitsa River Catchment. Embedded within a transdisciplinary landscape restoration project (the Tsitsa Project), this study combines a diagram‐based analysis of local actor mental models with a narrative‐based analysis of participant observation data. Challenges, opportunities, and general principles are identified. Limited accessibility, resource constraints, and the legacy of centralised governance impede participation. The research identifies the critical role of local actors, particularly those with deep community ties, in driving inclusive processes. Building trust through face‐to‐face interactions and culturally appropriate communication strategies also emerged as vital elements for success. This research can inform participatory projects and interventions in rural and developing world contexts.

Multidisciplinary cancer disease classification using adaptive FL in healthcare industry 5.0

Emerging Industry 5.0 designs promote artificial intelligence services and data-driven applications across multiple places with varying ownership that need special data protection and privacy considerations to prevent the disclosure of private information to outsiders. Due to this, federated learning offers a method for improving machine-learning models without accessing the train data at a single manufacturing facility. We provide a self-adaptive framework for federated machine learning of healthcare intelligent systems in this research. Our method takes into account the participating parties at various levels of healthcare ecosystem abstraction. Each hospital trains its local model internally in a self-adaptive style and transmits it to the centralized server for universal model optimization and communication cycle reduction. To represent a multi-task optimization issue, we split the dataset into as many subsets as devices. Each device selects the most advantageous subset for every local iteration of the model. On a training dataset, our initial study demonstrates the algorithm's ability to converge various hospital and device counts. By merging a federated machine-learning approach with advanced deep machine-learning models, we can simply and accurately predict multidisciplinary cancer diseases in the human body. Furthermore, in the smart healthcare industry 5.0, the results of federated machine learning approaches are used to validate multidisciplinary cancer disease prediction. The proposed adaptive federated machine learning methodology achieved 90.0%, while the conventional federated learning approach achieved 87.30%, both of which were higher than the previous state-of-the-art methodologies for cancer disease prediction in the smart healthcare industry 5.0.

Federated regressive learning: Adaptive weight updates through statistical information of clients

Federated learning is a method for training models in a distributed environment where each client utilizes its local dataset to train a model and shares it with a server to create a global model. This approach offers various advantages, such as eliminating the need for local clients to directly transfer their data to the server. However, it also has limitations, including the potential for limited global model performance due to imbalanced local datasets. In this study, we propose a new model that evaluates the quality of each client’s local dataset based on its probability distribution and quantity. Based on this evaluation, the contribution of locally trained models to the global model is adjusted accordingly. Unlike previous studies, our proposed method addresses the issue of low-quality local models degrading the performance of the global model. This allows for the improvement of the global model’s performance without the need for complex processes, such as adding specially designed loss functions to local models or employing data augmentation techniques to enhance the quality of the dataset. Furthermore, the method of calculating the contribution of local models proposed in this study allows for the quantitative assessment of the performance of clients participating in federated learning based on their data collection capabilities, making it an industry-friendly approach. We conducted experiments using various benchmark datasets, simulating different scenarios that could occur in the real world. In most scenarios, except a few, our method showed superior performance on all evaluation metrics compared to existing studies reviewed in the paper.

Data-free knowledge distillation via generator-free data generation for Non-IID federated learning

Data heterogeneity (Non-IID) on Federated Learning (FL) is currently a widely publicized problem, which leads to local model drift and performance degradation. Because of the advantage of knowledge distillation, it has been explored in some recent work to refine global models. However, these approaches rely on a proxy dataset or a data generator. First, in many FL scenarios, proxy dataset do not necessarily exist on the server. Second, the quality of data generated by the generator is unstable and the generator depends on the computing resources of the server. In this work, we propose a novel data-Free knowledge distillation approach via generator-Free Data Generation for Non-IID FL, dubbed as FedF2DG. Specifically, FedF2DG requires only local models to generate pseudo datasets for each client, and can generate hard samples by adding an additional regularization term that exploit disagreements between local model and global model. Meanwhile, FedF2DG enables flexible utilization of computational resources by generating pseudo dataset locally or on the server. And to address the label distribution shift in Non-IID FL, we propose a Data Generation Principle that can adaptively control the label distribution and number of pseudo dataset based on client current state, and this allows for the extraction of more client knowledge. Then knowledge distillation is performed to transfer the knowledge in local models to the global model. Extensive experiments demonstrate that our proposed method significantly outperforms the state-of-the-art FL methods and can serve as plugin for existing Federated Learning methds such as FedAvg, FedProx, etc, and improve their performance.

On an efficient global/local stochastic methodology for accurate stress analysis, failure prediction and damage tolerance of laminated composites

The quantification of uncertainties in the mechanical response of composite structures can be a computationally demanding task. This is due both to the number of uncertain parameters in a real study case and the complexity of the model to be analysed. In this paper, an efficient global/local approach to estimate the uncertainties of the quantities of interest in specific regions of interest with limited computational effort is proposed. This is achieved by refining only locally the model taking advantage of Refined Structural Theories. At the same time, since the variance of the uncertain parameters is usually relatively small, the stochastic analysis is dealt with a sensitivity study carried out both in the global and in the local model. In this way, it is possible to assess the influence of global and local uncertain parameters in the same submodeling analysis. The methodology presented is applied to several study cases of interest. The results focus on obtaining probabilistic distributions of the stress field that can be later used in failure criteria to evaluate the subsequent distribution of the failure index. Furthermore, a damage tolerance study case is investigated, showing good correlation with the reference Monte Carlo simulations.

Spatial clustering based on geographically weighted multivariate generalized gamma regression

Geographically Weighted Regression (GWR) is one of the local statistical models that can capture the effects of spatial heterogeneity. This model can be used for both univariate and multivariate responses. However, it should be noted that GWR models require the assumption of error normality. To overcome this problem, we propose a GWR model for generalized gamma distributed responses that can capture the phenomenon of some special continuous distributions. The proposed model is known as Geographically Weighted Multivariate Generalized Gamma Regression (GWMGGR). Parameter estimation is performed using the Maximum Likelihood Estimation (MLE) method optimized with the Bernt-Hall-Hall-Haussman (BHHH) algorithm. To determine the significance of the spatial heterogeneity effect, a hypothesis test was conducted using the Maximum Likelihood Ratio Test (MLRT) approach. We made a spatial cluster based on the estimated model parameters for each response using the k-means clustering method to interpret the obtained results. Some highlights of the proposed method are:•A new model for GWR with multivariate generalized gamma distributed responses to overcome the assumption of normally distributed errors.•Goodness of fit test to test the spatial effects in GWMGGR model.•Spatial clustering of districts/cities in Central Java based on three dimensions of educational indicators.

UTR: A UNet-like transformer for efficient unsupervised medical image registration

The existing medical image registration algorithms have the problem of low registration accuracy when processing large deformation medical images. In order to improve registration performance and utilize the global context extraction ability of Transformers without causing high computational complexity, a UNet-like Transformer model combining CNN and Transformer was constructed for 3D medical image registration tasks. We use the Efficient Global Local Attention (EGLA) mechanism to construct a Transformer encoder to further address the difficulty of modeling long-distance dependencies in existing medical image registration networks. We leverage the local modeling capabilities of CNN and the long-distance information capture capabilities of Transformer to achieve high-precision registration. The algorithm has undergone detailed validation experiments on two public datasets. The qualitative and quantitative registration results validate the effectiveness of the proposed model.

BR-FEEL: A backdoor resilient approach for federated edge learning with fragment-sharing

In the resource-constrained federated edge learning (FEEL) systems, fragment-sharing is an efficient approach for multiple clients to cooperatively train a giant model with billions of parameters. Compared with the classical federated learning schemes where the local model is fully trained and exchanged by each client, the fragment-sharing only requires each client to optionally choose a parameter-fragment to train and share, according to its storage, computing, and networking abilities. However, when the full model is no longer delivered in fragment-sharing, the backdoor attacks hidden behind the fragments become harder to be detected, which introduces formidable challenge for the security of FEEL systems. In this paper, we firstly show that the existing fragment-sharing works suffer a lot from the backdoor attacks. Then, a Backdoor-Resilient approach, named BR-FEEL, is introduced to defend against the potential backdoor attacks. Specifically, a twin model is built by each benign client to integrate the parameter-fragments from others. A knowledge distillation process is designed on each client to transfer the clean knowledge from its twin model to local model. With the twin model and knowledge distillation process, our BR-FEEL approach makes sure that the local models of the benign clients will not be backdoored. Experiments on CIFAR-10 and GTSRB datasets with MobileNetV2 and ResNet-34 are conducted. The numerical results demonstrate the efficacy of BR-FEEL on reducing attack success rates by over 90% compared to other baselines under various attack methods.

An Intelligent Model of Anomaly Detection and Anomaly Localization in Images Using Hybrid Heuristic Adaptive Multiscale Attention-Based DenseNet and Cascaded Variational Autoencoder

Anomaly detection and localization become necessary since humans are unable to spot visual frauds. Since it differs from the normal representation of shapes, colors, textures, size, and location, it does not have the potential to be employed in such practical applications. Hence, the automatic anomaly detection and localization framework is suggested for supporting situations like industrial systems, healthcare systems and so on. The detection and localization are the most intriguing and challenging concerns in image processing. In general, anomalies can manifest themselves in diverse ways. Detection is the process of providing information about whether an image contains modifications or not. On the other hand, localization is defined as locating the anomalies in images. Several traditional models have been deployed, but they still lack such constraints to evade performance enhancement. To sort out this issue, a novel adaptive method is recommended for anomaly detection and localization in images. Initially, the source images are acquired from standard public data sources. Once the required images are collected, such abnormal or anomalous activities are detected by adopting the new method known as Adaptive Multiscale Attention-based DenseNet (AMA-DeNet), where the hyperparameters are tuned by a hybrid algorithm known as the Hybrid Chimp Grasshopper Optimization Algorithm (HCGOA). Subsequently, the final stage is localizing the anomalies in the detected images. This is to be accomplished by using the Cascaded Variational Autoencoder (CVA), in which parameters are optimally chosen by HCGOA. After implementation, the experimental results are computed across diverse validating measures. Hence, the extensive results elucidate that the recommended model contains the potential to detect the abnormalities in the images accurately, thereby enhancing the efficiency of the new system.

Eukaryotic Chemotaxis under Periodic Stimulation Shows Temporal Gradient Dependence.

When cells of the social amoeba Dictyostelium discoideum are starved of nutrients they start to synthesize and secrete the chemical messenger and chemoattractant cyclic adenosine monophosphate (cAMP). This signal is relayed by other cells, resulting in the establishment of periodic waves. The cells aggregate through chemotaxis toward the center of these waves. We investigated the chemotactic response of individual cells to repeated exposure to waves of cAMP generated by a microfluidic device. For fast-moving waves (short period), the chemotactic ability of the cells was found to increase upon exposure to more waves, suggesting the development of a memory over several cycles. This effect was not significant for slow-moving waves (large period). We show that the experimental results are consistent with a local excitation global inhibition-based model, extended by including a component that rises and decays slowly and that is activated by the temporal gradient of cAMP concentration. The observed enhancement in chemotaxis is relevant to populations in the wild: once sustained, periodic waves of the chemoattractant are established, it is beneficial to cells to improve their chemotactic ability in order to reach the aggregation center sooner.

Evaluating Federated Learning Simulators: A Comparative Analysis of Horizontal and Vertical Approaches.

Federated learning (FL) is a decentralized machine learning approach whereby each device is allowed to train local models, eliminating the requirement for centralized data collecting and ensuring data privacy. Unlike typical typical centralized machine learning, collaborative model training in FL involves aggregating updates from various devices without sending raw data. This ensures data privacy and security while collecting a collective learning from distributed data sources. These devices in FL models exhibit high efficacy in terms of privacy protection, scalability, and robustness, which is contingent upon the success of communication and collaboration. This paper explore the various topologies of both decentralized or centralized in the context of FL. In this respect, we investigated and explored in detail the evaluation of four widly used end-to-end FL frameworks: FedML, Flower, Flute, and PySyft. We specifically focused on vertical and horizontal FL systems using a logistic regression model that aggregated by the FedAvg algorithm. specifically, we conducted experiments on two images datasets, MNIST and Fashion-MNIST, to evaluate their efficiency and performance. Our paper provides initial findings on how to effectively combine horizontal and vertical solutions to address common difficulties, such as managing model synchronization and communication overhead. Our research indicates the trade-offs that exist in the performance of several simulation frameworks for federated learning.

Adapting the segment anything model for multi-modal retinal anomaly detection and localization

The fusion of optical coherence tomography (OCT) and fundus modality information can provide a comprehensive diagnosis for retinal artery occlusion (RAO) disease, where OCT provides the cross-sectional examination of the fundus image. Given multi-modal retinal images, an anomaly diagnosis model can discriminate RAO without the need for real diseased samples. Despite this, previous studies have only focused on single-modal diagnosis, because of: 1) the lack of paired modality samples; and 2) the significant imaging differences, which make the fusion difficult with small-scale medical data. In this paper, we describe how we first built a multi-modal RAO dataset including both OCT and fundus modalities, which supports both the anomaly detection and localization tasks with pixel-level annotation. Motivated by the powerful generalization ability of the recent visual foundation model known as the Segment Anything Model (SAM), we adapted it for our task considering the small-scale property of retinal samples. Specifically, a modality-shared decoder with task-specific tokens is introduced to make SAM support the multi-modal image setting, which includes a mask token for the anomaly localization task at the pixel level and a fusion token for the anomaly detection task at the case level. Since SAM has little medical knowledge and lacks the learning of the “normal” concept, it is infeasible to localize RAO anomalies in the zero-shot manner. To integrate expert retinal knowledge while keeping the general segmentation knowledge, general anomaly simulation for both modalities and a low-level prompt-tuning strategy are introduced. The experiments conducted in this study show that the adapted model can surpass the state-of-the-art model by a large margin. This study sets the first benchmark for the multi-modal anomaly detection and localization tasks in the medical community. The code is available at https://github.com/Jingtao-Li-CVer/MMRAD.