Task In Applications Research Articles

Ultrahigh Energy Storage Performance in BiFeO3-Based Lead-Free Ceramics via Tuning Structural Homogeneity and Domain Engineering Strategies.

Lead-free ceramic-based dielectric capacitors are critical in electronics and environmental safety. Nevertheless, developing ideal lead-free ceramics with excellent energy storage properties remains a challenging task for practical applications. Herein, the enhanced relaxation behavior and increased breakdown electric field are utilized to realize the high energy storage behavior of (0.7-x)BiFeO3-0.3BaTiO3-x(Na0.7Sm0.1)NbO3(BF-BT-xNSN) ceramics. The introduction of complex perovskite compound NSN facilitates the transition of domains to nanodomains and improves the homogeneity of the microstructure, which results in the enhancement of the relaxation properties and breakdown electric field. Hence, the BF-BT-0.2NSN ceramic can achieve an ultrahigh energy storage density of 13.1 J/cm3 under an electric field of 650 kV/cm. Moreover, the designed BF-BT-0.2NSN ceramic achieves remarkable thermal stability (20-100 °C), frequency stability (1-100 Hz), and excellent charge/discharge performance. This study develops an idea of dielectric capacitor design and reveals the remarkable potential of BiFeO3-based dielectric ceramics within the realm of energy storage applications.

Interpretable machine learning for time-to-event prediction in medicine and healthcare.

Time-to-event prediction, e.g. cancer survival analysis or hospital length of stay, is a highly prominent machine learning task in medical and healthcare applications. However, only a few interpretable machine learning methods comply with its challenges. To facilitate a comprehensive explanatory analysis of survival models, we formally introduce time-dependent feature effects and global feature importance explanations. We show how post-hoc interpretation methods allow for finding biases in AI systems predicting length of stay using a novel multi-modal dataset created from 1235 X-ray images with textual radiology reports annotated by human experts. Moreover, we evaluate cancer survival models beyond predictive performance to include the importance of multi-omics feature groups based on a large-scale benchmark comprising 11 datasets from The Cancer Genome Atlas (TCGA). Model developers can use the proposed methods to debug and improve machine learning algorithms, while physicians can discover disease biomarkers and assess their significance. We contribute open data and code resources to facilitate future work in the emerging research direction of explainable survival analysis.

VECTOR: Velocity-Enhanced GRU Neural Network for Real-Time 3D UAV Trajectory Prediction

This paper addresses the challenge of predicting 3D trajectories for Unmanned Aerial Vehicles (UAVs) in real-time, a critical task for applications like aerial surveillance and defense. Current prediction models primarily leverage only position data, which may not provide the most accurate forecasts for UAV movements and usually fail outside the position domain used in the training phase. Our research identifies a gap in utilizing velocity estimates and first-order dynamics to better capture the dynamics and enhance prediction accuracy and generalizability in any position domain. To bridge this gap, we introduce a trajectory prediction scheme using sequence-based neural networks with Gated Recurrent Units (GRUs) to forecast future velocity and positions based on historical velocity estimates instead of position measurements. This approach is designed to improve the predictive capabilities over traditional methods that rely solely on recurrent neural networks (RNNs) or transformers, which can struggle with scalability in this context. Our methodology employs both synthetic and real-world 3D UAV trajectory data, incorporating diverse patterns of agility, curvature, and speed. Synthetic data are generated using the Gazebo robotics simulator and PX4 Autopilot, while real-world data are sourced from the UZH-FPV and Mid-Air drone racing datasets. We train the GRU-based models on drone 3D position and velocity samples to capture the dynamics of UAV movements effectively. Quantitatively, the proposed GRU-based prediction algorithm demonstrates superior performance, achieving a mean square error (MSE) ranging from 2×10−8 to 2×10−7. This performance outstrips existing state-of-the-art RNN models. Overall, our findings confirm the effectiveness of incorporating velocity data in improving the accuracy of UAV trajectory predictions across both synthetic and real-world scenarios, in and out of position data distributions. Finally, we open-source our 5000 trajectories dataset and a ROS2 package to facilitate the integration with existing ROS-based UAV systems.

Region-Focusing Data Augmentation via Salient Region Activation and Bitplane Recombination for Target Detection

As the performance of a convolutional neural network is logarithmically proportional to the amount of training data, data augmentation has attracted increasing attention in recent years. Although the current data augmentation methods are efficient because they force the network to learn multiple parts of a given training image through occlusion or re-editing, most of them can damage the internal structures of targets and ultimately affect the results of subsequent application tasks. To this end, region-focusing data augmentation via salient region activation and bitplane recombination for the target detection of optical satellite images is proposed in this paper to solve the problem of internal structure loss in data augmentation. More specifically, to boost the utilization of the positive regions and typical negative regions, a new surroundedness-based strategy for salient region activation is proposed, through which new samples with meaningful focusing regions can be generated. And to generate new samples of the focusing regions, a region-based strategy for bitplane recombination is also proposed, through which internal structures of the focusing regions can be reserved. Thus, a multiplied effect of data augmentation by the two strategies can be achieved. In addition, this is the first time that data augmentation has been examined from the perspective of meaningful focusing regions, rather than the whole sample image. Experiments on target detection with public datasets have demonstrated the effectiveness of this proposed method, especially for small targets.

In vitro screening of potential echinochrome delivery systems for the treatment of eye diseases

An important task of topical application of medicines in the treatment of eyes is to achieve a compromise between their effectiveness and safety. The development of new multifunctional local ophthalmic drug delivery systems and in vitro screening of potential medicinal eye products are key areas in solving this problem. In this study, primary in vitro screening of the effect of echinochrome (Ech), the carrageenan complex of echinochrome (CRG/Ech) and its liposomal form (CRG/Ech-Lip) was performed on cultured epithelial cells of the outer shell of the eyeball: conjunctival epithelial cells (Chang Conjunctiva, Clone 1-5c-4) and corneal epithelium human (HCE). The cell viability was assessed by their morphology and metabolic activity using light microscopy and MTT test methods. The direct dependence of the intensity of the cytotoxic effect of Ech on its concentration in the nutrient medium, the form of use, the cellular test system and the incubation time of cells was revealed. Ech in the form of an alcoholic solution in its final concentration of 0.1 mg/ml of the nutrient medium exhibits pronounced cytoxicity against both cellular test systems. The same final concentration of Ech in the nutrient medium, but already as part of the carrageenan complex of echinochrome (CRG/Ech), turned out to be critical only for the viability of corneal epithelial cells, the survival rate of conjunctival cells under these conditions was about 50 %. A high biocompatibility of the liposomal form of the carrageenan complex of echinochrome (CRG/Ech-Lip) with cells of both test systems and a stimulating cytoprotective effect against the cells of the conjunctiva epithelium was revealed.

3D Non-separable Moment Invariants and Their Use in Neural Networks

Recognition of 3D objects is an important task in many bio-medical and industrial applications. The recognition algorithms should work regardless of a particular orientation of the object in the space. In this paper, we introduce new 3D rotation moment invariants, which are composed of non-separable Appell moments. We show that non-separable moments may outperform the separable ones in terms of recognition power and robustness thanks to a better distribution of their zero surfaces over the image space. We test the numerical properties and discrimination power of the proposed invariants on three real datasets—MRI images of human brain, 3D scans of statues, and confocal microscope images of worms. We show the robustness to resampling errors improved more than twice and the recognition rate increased by 2–10 % comparing to most common descriptors. In the last section, we show how these invariants can be used in state-of-the-art neural networks for image recognition. The proposed H-NeXtA architecture improved the recognition rate by 2–5 % over the current networks.

Integration of Leap Motion sensor with camera for better performance in hand tracking

<p>In this paper, we propose a framework for hand tracking in human-computer interaction applications. Leap Motion is used today as a popular interface in virtual reality and computer games. In this study, we evaluated the merits and drawbacks of this device. The limitations of this device restrict the user’s free movement. The purpose of this study is to find an optimal solution for using Leap Motion. We propose a framework to estimate the pose of the hand in a bigger space around Leap Motion. Our framework uses an integration of Leap Motion with a camera that are placed in two different places to capture the information of the hand from various views. The experiments are designed based on the common tasks in the human-computer interaction applications. The finding of this study demonstrates that the proposed framework increases the precise interaction space.</p>

Helico-conical vector beams for intensity and polarization 3D light shaping

While vector beams offer an intriguing way to structure optical beams and enhance light-based technologies across many fields, their generation remains a challenging task in practical applications. Disclosing an unprecedented manipulation of light at the subwavelength scale, metaoptics have inspired smart and efficient solutions for spatially variant polarization structuring. Concurrently, the generalization of non-separability in polarization and phase manipulation extends the vectorial nature beyond standard vector vortices. In this work, we present the design and test of dual-functional metasurfaces for the compact generation of a new type of vector beam, so-called helico-conical vector beam, providing an inhomogeneous polarization pattern over customizable one-arm or two-arm 3D spirals of light. These devices pave the way to integrated optical architectures for dynamic optical manipulation and trapping in many fields, from optofluidics to quantum computing.

Disentangled Representation Learning.

Disentangled Representation Learning (DRL) aims to learn a model capable of identifying and disentangling the underlying factors hidden in the observable data in representation form. The process of separating underlying factors of variation into variables with semantic meaning benefits in learning explainable representations of data, which imitates the meaningful understanding process of humans when observing an object or relation. As a general learning strategy, DRL has demonstrated its power in improving the model explainability, controlability, robustness, as well as generalization capacity in a wide range of scenarios such as computer vision, natural language processing, and data mining. In this article, we comprehensively investigate DRL from various aspects including motivations, definitions, methodologies, evaluations, applications, and model designs. We first present two well-recognized definitions, i.e., Intuitive Definition and Group Theory Definition for disentangled representation learning. We further categorize the methodologies for DRL into four groups from the following perspectives, the model type, representation structure, supervision signal, and independence assumption. We also analyze principles to design different DRL models that may benefit different tasks in practical applications. Finally, we point out challenges in DRL as well as potential research directions deserving future investigations. We believe this work may provide insights for promoting the DRL research in the community.

Predefined-Time Convergent Kinematic Control of Robotic Manipulators With Unknown Models Based on Hybrid Neural Dynamics and Human Behaviors.

This article proposes a model-free kinematic control method with predefined-time convergence for robotic manipulators with unknown models. The predefined-time convergence property guarantees that the regulation task can be finished by robotic manipulators in a preset time, in spite of the initial state of manipulators. This feature will facilitate the scheduling of a series of tasks in industrial applications. To this end, a varying-parameter predefined-time convergent zeroing neural dynamics (ZND) model is first proposed and employed to solve the regulation problem. As well as the primary task, a conventional ZND model is utilized to achieve the avoidance of obstacle. The stability of the proposed controller is analyzed based on the Lyapunov stability theory. For the sake of dealing with the unknown kinematic model of robotic manipulators, gradient neural dynamics (GND) models are exploited to adapt the Jacobian matrices just relying on the control signal and sensory output, which enables us to control robotic manipulators in a model-free manner. Finally, the efficacy and merits of the proposed control method are verified by simulations and experiments, including a comparison with the existing method.

Adaptive Annotation Correlation Based Multi-Annotation Learning for Calibrated Medical Image Segmentation.

Medical image segmentation is a fundamental task in many clinical applications, yet current automated segmentation methods rely heavily on manual annotations, which are inherently subjective and prone to annotation bias. Recently, modeling annotator preference has garnered great interest, and several methods have been proposed in the past two years. However, the existing methods completely ignore the potential correlation between annotations, such as complementary and discriminative information. In this work, the Adaptive annotation CorrelaTion based multI-annOtation LearNing (ACTION) method is proposed for calibrated medical image segmentation. ACTION employs consensus feature learning and dynamic adaptive weighting to leverage complementary information across annotations and emphasize discriminative information within each annotation based on their correlations, respectively. Meanwhile, memory accumulation-replay is proposed to accumulate the prior knowledge and integrate it into the model to enable the model to accommodate the multi-annotation setting. Two medical image benchmarks with different modalities are utilized to evaluate the performance of ACTION, and extensive experimental results demonstrate that it achieves superior performance compared to several state-of-the-art methods.

Impact of random variations in the sampling period on the stability of embedded control systems

Abstract Control of embedded systems is a key task in many industrial and technological applications, and their performance and stability are critically dependent on the ability to effectively implement control algorithms within limited computational and time resources. One of the main challenges in this area is the need to process data with a random variable that comes to processing randomly and with variable periodicity. This variability in the sampling period can cause a decrease in the quality of the control, which threatens the stability of the entire system. The article examines the limits to which this variability can be tolerated without losing the stability of the controlled system. The study includes an analysis of systems dynamics with a variable sampling period and determining the necessary period of system discretization. It is also essential to consider the impact of the random component of the sampling period on the system’s transient response. The proposal is based on determining the permissible limits of the period change based on analysis and simulation. The result of this research provides recommendations for the design and implementation of control systems that can maintain high-quality regulation even with unpredictable changes in sampling periodicity. The present article also opens space for further research in the field of control of systems with variable periodicity, especially regarding their practical implementation in real embedded devices.

Dual-Pixel Raindrop Removal.

Removing raindrops in images has been addressed as a significant task for various computer vision applications. In this paper, we propose the first method using a dual-pixel (DP) sensor to better address raindrop removal. Our key observation is that raindrops attached to a glass window yield noticeable disparities in DP's left-half and right-half images, while almost no disparity exists for in-focus backgrounds. Therefore, the DP disparities can be utilized for robust raindrop detection. The DP disparities also bring the advantage that the occluded background regions by raindrops are slightly shifted between the left-half and the right-half images. Therefore, fusing the information from the left-half and the right-half images can lead to more accurate background texture recovery. Based on the above motivation, we propose a DP Raindrop Removal Network (DPRRN) consisting of DP raindrop detection and DP fused raindrop removal. To efficiently generate a large amount of training data, we also propose a novel pipeline to add synthetic raindrops to real-world background DP images. Experimental results on constructed synthetic and real-world datasets demonstrate that our DPRRN outperforms existing state-of-the-art methods, especially showing better robustness to real-world situations.

The Historical Evolution and Significance of Multiple Sequence Alignment in Molecular Structure and Function Prediction.

Multiple sequence alignment (MSA) has evolved into a fundamental tool in the biological sciences, playing a pivotal role in predicting molecular structures and functions. With broad applications in protein and nucleic acid modeling, MSAs continue to underpin advancements across a range of disciplines. MSAs are not only foundational for traditional sequence comparison techniques but also increasingly important in the context of artificial intelligence (AI)-driven advancements. Recent breakthroughs in AI, particularly in protein and nucleic acid structure prediction, rely heavily on the accuracy and efficiency of MSAs to enhance remote homology detection and guide spatial restraints. This review traces the historical evolution of MSA, highlighting its significance in molecular structure and function prediction. We cover the methodologies used for protein monomers, protein complexes, and RNA, while also exploring emerging AI-based alternatives, such as protein language models, as complementary or replacement approaches to traditional MSAs in application tasks. By discussing the strengths, limitations, and applications of these methods, this review aims to provide researchers with valuable insights into MSA's evolving role, equipping them to make informed decisions in structural prediction research.

Adaptive target localization under uncertainty using Multi-Agent Deep Reinforcement Learning with knowledge transfer

Target localization is a critical task in sensitive applications, where multiple sensing agents communicate and collaborate to identify the target location based on sensor readings. Existing approaches investigated the use of Multi-Agent Deep Reinforcement Learning (MADRL) to tackle target localization. Nevertheless, these methods do not consider practical uncertainties, like false alarms when the target does not exist or when it is unreachable due to environmental complexities. To address these drawbacks, this work proposes a novel MADRL-based method for target localization in uncertain environments. The proposed MADRL method employs Proximal Policy Optimization to optimize the decision-making of sensing agents, which is represented in the form of an actor–critic structure using Convolutional Neural Networks. The observations of the agents are designed in an optimized manner to capture essential information in the environment, and a team-based reward functions is proposed to produce cooperative agents. The MADRL method covers three action dimensionalities that control the agents’ mobility to search the area for the target, detect its existence, and determine its reachability. Using the concept of Transfer Learning, a Deep Learning model builds on the knowledge from the MADRL model to accurately estimating the target location if it is unreachable, resulting in shared representations between the models for faster learning and lower computational complexity. Collectively, the final combined model is capable of searching for the target, determining its existence and reachability, and estimating its location accurately. The proposed method is tested using a radioactive target localization environment and benchmarked against existing methods, showing its efficacy.

On the feasibility of an ensemble multi-fidelity neural network for fast data assimilation for subsurface flow in porous media

Uncertainty quantification (UQ) of the reservoir heterogeneity is essential to predict fluid flow behavior in subsurface formations accurately, and the task is often accomplished by integrating high-fidelity forward physics simulators with iterative data assimilation methods, and such workflows are usually computationally expensive due to the iterative nature and the prohibitive cost of physics simulations.In this work, we develop a new Ensemble Multi-Fidelity Neural Network (EMF-Net) to mitigate the efficiency bottleneck of UQ. EMF-Net directly infers the uncertain variables (e.g., permeability) via the sparse observation of the state variables (e.g., pressure). By leveraging the regression capability of deep neural networks, EMF-Net directly learns the nonlinear mapping from the innovation vector to the inferred update vector without the hypothesis of a linear mapping. At the training phase, high-fidelity data computed by a physics simulator (fh) and low-fidelity data computed by a forward proxy model (fl) are used to train the EMF-Net at two different stages, respectively. At the inference phase, we first adopt the 1-stage EMF-Net to infer the update vector for the initial ensembles with a global tuning and then efficiently update the innovation vector by fl. As an option, we further refine the inference, via a 2-stage EMF-Net, which captures the locality and enhances consistency with the observation data. We demonstrate that EMF-Net can reach equivalent inference accuracy compared to classic data assimilation algorithms like ES-MDA, in addition to decreasing the CPU time. Therefore, the accuracy and efficiency make it an attractive alternative for scalable real-time history-matching tasks in field-scale subsurface engineering applications.

Assessing vocabulary acquisition using a fast-mapping task in an Android application: A pilot study

Purpose The aim of this study was to explore whether a fast mapping task embedded in an Android application (FastMApp) is a valid tool to assess referent selection abilities in Spanish-speaking children aged between 18 and 30 months. Traditional assessment tools for lexical development use static quantitative methods that assign children a final score to represent their overall vocabulary level. These methods fail to provide insights into the learning process, despite their potential relevance for clinical and educational purposes. Method Sixty Spanish-speaking children participated in this study. They completed the FastMApp (a 22-trials’ fast mapping noun task including 4- and 5-item trials, with one unknown object), and their caregivers rated their child’s vocabulary on a parent-rated vocabulary inventory measure. Result The data show a high percentage of responses to the task, indicating that the children were actively complied with the task. The scores for known labels are significantly higher compared to unknown labels, and the scores for 4-item trials are significantly higher compared to 5-item trials. We observed a strong and significant relationship between the scores in this task and the scores on the parent-rated vocabulary inventory measure. Conclusion The results suggest that FastMApp is suitable for assessing early vocabulary acquisition in Spanish-speaking children.

Deep Reinforcement Learning for Dynamic Task Scheduling in Edge-Cloud Environments

With The advent of the Internet of Things (IoT) and its use cases there is a necessity for improved latency which has led to edgecomputing technologies. IoT applications need a cloud environment and appropriate scheduling based on the underlying requirements of a given workload. Due to the mobility nature of IoT devices and resource constraints and resource heterogeneity, IoT application tasks need more efficient scheduling which is a challenging problem. The existing conventional and deep learning scheduling techniques have limitations such as lack of adaptability, issues with synchronous nature and inability to deal with temporal patterns in the workloads. To address these issues, we proposed a learning-based framework known as the Deep Reinforcement Learning Framework (DRLF). This is designed in such a way that it exploits Deep Reinforcement Learning (DRL) with underlying mechanisms and enhanced deep network architecture based on Recurrent Neural Network (RNN). We also proposed an algorithm named Reinforcement Learning Dynamic Scheduling (RLbDS) which exploits different hyperparameters and DRL-based decision-making for efficient scheduling. Real-time traces of edge-cloud infrastructure are used for empirical study. We implemented our framework by defining new classes for CloudSim and iFogSim simulation frameworks. Our empirical study has revealed that RLbDS out performs many existing scheduling methods.

Engagement with task context of applications tasks: student performance and teacher beliefs

Engagement with task context of applications tasks: student performance and teacher beliefs

Dual-dimensional contrastive learning for incomplete multi-view clustering

Incomplete multi-view clustering (IMVC) is a critical task in real-world applications, where missing data in some views can severely limit the ability to leverage complementary information across views. This issue leads to incomplete sample representations, hindering model performance. Current contrastive learning methods for IMVC exacerbate the problem by directly constructing data pairs from incomplete samples, ignoring essential information and resulting in class collisions, where samples from different classes are incorrectly grouped together due to a lack of label guidance. These challenges are particularly detrimental in fields like recommendation systems and bioinformatics, where accurate clustering of high-dimensional and incomplete data is essential for decision-making. To address these issues, we propose Dual-dimensional Contrastive Learning (DCL), an online IMVC model that fills missing values through multi-view consistency transfer, enabling simultaneous clustering and representation learning via instance-level and cluster-level contrastive learning in both row and column spaces. DCL mitigates class collision issues by generating high-confidence pseudo-labels and using an optimal transport matrix, significantly improving clustering accuracy. Extensive experiments demonstrate that DCL achieves state-of-the-art results across five datasets. The code is available at https://github.com/2251821381/DCL.