検索結果 - "weight loss"

結果の絞り込み
  1. 981

    A Simple and Effective KAN-Based Architecture for Accurate Battery RUL Prediction 著者: Guangzai Ye, Li Feng, Jianlan Guo, Yuqiang Chen, Shufei Li

    出版事項 2025-01-01

    Accurately estimating a lithium-ion battery’s Remaining Useful Life (RUL) is crucial for ensuring the safety and reliability of battery management systems. However, the performance of emerging architectures, such as Kolmogorov-Arnold Networks (KANs), is often hindered by the significant n...

    詳細記述

    “…It also incorporates a channel-independent KANs structure with a regularized weighted loss function to handle variable-specific degradation patterns. …”
    全文の入手
    論文
  2. 982

    High performance sulfur-containing copolyesters from bio-sourced aromatic monomers 著者: Lesly Dasilva Wandji Djouonkep, Zhengzai Cheng, William Mawuko Kodjo Siegu, Xiong Jing, Jun Chen, Elvis Kwame Adom, Abubakar Muaz, Mario Gauthier

    出版事項 2022-01-01

    In this investigation, a series of novel random bio-based thiophene–aromatic copolyesters, including thiophene and phenyl units, were successfully prepared from dimethyl 2,5-thiophenedicarboxylate, dimethyl 2,5-dimethoxyterephthalate, and the aliphatic diols ethylene glycol, 1,3-propanediol, 1,4-but...

    詳細記述

    “…After 28 weeks of incubation in humid soil, weight losses of up to 7.2% were observed. Considering their good mechanical properties, thermal stability and biodegradability, these renewable sulfur-containing copolyesters have great potential to replace petroleum-based commercial poly(ethylene terephthalate) in the food packaging industry, which is helpful to implement carbon neutrality and sustainable development in the polymer industry.…”
    全文の入手
    論文
  3. 983

    Deep learning-based diffusion MRI tractography: Integrating spatial and anatomical information 著者: Yiqiong Yang, Yitian Yuan, Baoxing Ren, Ye Wu, Yanqiu Feng, Xinyuan Zhang

    出版事項 2025-08-01

    Diffusion MRI tractography technique enables non-invasive visualization of the white matter pathways in the brain. It plays a crucial role in neuroscience and clinical fields by facilitating the study of brain connectivity and neurological disorders. However, the accuracy of reconstructed tractogram...

    詳細記述

    “…Additionally, we employ a weighted loss function to address fiber class imbalance encountered during training. …”
    全文の入手
    論文
  4. 984

    Domain adaptive deep possibilistic clustering for EEG-based emotion recognition 著者: Yufang Dan, Yufang Dan, Yufang Dan, Yufang Dan, Yufang Dan, Qun Li, Xianhua Wang, Di Zhou

    出版事項 2025-07-01

    Emotion recognition based on electroencephalogram (EEG) faces substantial challenges. The variability of neural signals among different subjects and the scarcity of labeled data pose obstacles to the generalization ability of traditional domain adaptation (DA) methods. Existing approaches, especiall...

    詳細記述

    “…Moreover, the DADPc incorporates adaptive weighted loss and memory bank strategies to enhance the reliability of pseudo-labels and cross-domain alignment. …”
    全文の入手
    論文
  5. 985

    Optimizing the acid resistance of concrete with granulated blast-furnace slag 著者: Luca-Alexander Kempf, Rolf Breitenbücher, Christian Gerten, Andreas Ehrenberg

    出版事項 2022-03-01

    Concrete structures exposed to high levels of chemical attacks are assigned to exposure class XA3, which recommends separate concrete protection or a special expert solution to ensure durability. Due to the partial substitution of Portland cement by blast-furnace slag, an increased resistance to aci...

    詳細記述

    “…After 13 weeks of storing concrete specimens in sulfuric acid (H2SO4, pH 3.5), reduced damage depths and lower weight losses were observed compared to conventional binder compositions. …”
    全文の入手
    論文
  6. 986

    Optimizing 5G resource allocation with attention-based CNN-BiLSTM and squeeze-and-excitation architecture 著者: Anfal Musadaq Rayyis, Mohammad Maftoun, Maryam Khademi, Emrah Arslan, Silvia Gaftandzhieva

    出版事項 2025-07-01

    IntroductionThe swift advancement of computational capabilities has rendered deep learning indispensable for tackling intricate challenges. In 5G networks, efficient resource allocation is crucial for optimizing performance and minimizing latency. Traditional machine learning models struggle to capt...

    詳細記述

    “…Traditional machine learning models struggle to capture intricate temporal dependencies and handle imbalanced data distributions, limiting their effectiveness in real-world applications.MethodsTo overcome these limitations, this study presents an innovative deep learning-based framework that combines a convolutional layer with squeeze-and-excitation block, bidirectional long short-term memory, and a self-attention mechanism for resource allocation prediction. A custom weighted loss function addresses data imbalance, while Bayesian optimization fine-tunes hyperparameters.ResultsExperimental results demonstrate that the proposed model achieves state-of-the-art predictive accuracy, with a remarkably low Mean Absolute Error (MAE) of 0.0087, Mean Squared Error (MSE) of 0.0003, Root Mean Squared Error (RMSE) of 0.0161, Mean Squared Log Error (MSLE) of 0.0001, and Mean Absolute Percentage Error (MAPE) of 0.0194. …”
    全文の入手
    論文
  7. 987

    A Hybrid Quantum-Classical Approach for Multi-Class Skin Disease Classification Using a 4-Qubit Model 著者: Aravinda C V, Emerson Raja Joseph, Sultan Alasmari

    出版事項 2025-01-01

    Quantum machine learning (QML) presents a promising avenue for addressing complex classification challenges, yet its application in medical imaging remains largely unexplored. This work introduces a hybrid quantum-classical framework designed to classify skin diseases, Chickenpox, Measles, Monkeypox...

    詳細記述

    “…We employ a class-weighted loss function with weights <inline-formula> <tex-math notation="LaTeX">$w_{y_{i}} = [{3.0, 3.5, 0.6, 0.4}]$ </tex-math></inline-formula> to address this imbalance, ensuring balanced representation across all classes. …”
    全文の入手
    論文
  8. 988

    Ship Target Detection in SAR Images Based on Multiple Attention Mechanism and Cross-Scale Feature Fusion 著者: Yuwu Wang, Tieming Wu, Limin Guo, Yuhan Mo

    出版事項 2025-01-01

    Aiming to address the challenges of inefficient target detection in synthetic aperture radar (SAR) images caused by complex backgrounds, small ship targets, and significant scale variations, this article proposes a novel SAR ship target detection model, YOLO-SS, based on YOLOv10n. First, the method...

    詳細記述

    “…Furthermore, by incorporating the normalized Wasserstein distance (NWD) and combining it with the CIoU loss function, a CIoU-NWD weighted loss function is designed. This reduces the sensitivity of the CIoU loss function to positional offsets of small targets, with only a slight increase in computational and parameter costs, thereby further improving the detection accuracy of small targets. …”
    全文の入手
    論文
  9. 989

    Evaluating the role of training data origin for country-scale cropland mapping in data-scarce regions: A case study of Nigeria 著者: Joaquin Gajardo, Michele Volpi, Daniel Onwude, Thijs Defraeye

    出版事項 2025-08-01

    Cropland maps are essential for remote sensing-based agricultural monitoring, providing timely insights about agricultural development without requiring extensive field surveys. While machine learning enables large-scale mapping, it relies on geo-referenced ground-truth data, which is time-consuming...

    詳細記述

    “…Handling class imbalance was also critical, with weighted loss functions improving accuracy by up to 0.071 for the single-headed LSTM. …”
    全文の入手
    論文
  10. 990

    Referenceless 4D flow cardiovascular magnetic resonance with deep learning 著者: Chiara Trenti, Erik Ylipää, Tino Ebbers, Carl-Johan Carlhäll, Jan Engvall, Petter Dyverfeldt

    出版事項 2025-01-01

    ABSTRACT: Background: Despite its potential to improve the assessment of cardiovascular diseases, four-dimensional (4D) flow cardiovascular magnetic resonance (CMR) is hampered by long scan times. 4D flow CMR is conventionally acquired with three motion encodings and one reference encoding, as the...

    詳細記述

    “…Methods: A U-Net was trained with adversarial learning (U-NetADV) and with a velocity frequency-weighted loss function (U-NetVEL) to predict the reference encoding from the three motion encodings obtained with a non-symmetric velocity-encoding scheme. …”
    全文の入手
    論文
  11. 991

    Bismuth ferrite reinforced porous bioactive glass scaffolds: In vitro and antibacterial properties 著者: S. Amitha Banu, Sk Hasanur Rahaman, Khan Sharun, Merlin Mamachan, Swapan Kumar Maiti, Amarpal, Subhadip Bodhak, Vamsi Krishna Balla, Abhijit M. Pawde

    出版事項 2025-07-01

    Tissue engineering focuses on restoring damaged tissues by strategically integrating cells, bioactive factors, and scaffold materials. Despite significant advancements in biomaterials, developing an ideal scaffold for bone regeneration remains a major challenge. The porous scaffolds aim to provide a...

    詳細記述

    “…The BAG and BAG-BF porous scaffolds exhibited porosities of ∼74 % (BAG), ∼65 % (0.5 BAG-BF), and ∼64 % (1.5 BAG-BF), with post-sintering weight losses of 5 %, 2.5 %, and 5 %, respectively. All samples showed ∼50 % shrinkage. …”
    全文の入手
    論文
  12. 992

    Development of an interpretable machine learning model for frailty risk prediction in older adult care institutions: a mixed-methods, cross-sectional study in China 著者: Weizi Wu, Li Jing, Qing Peng, Peng Hua, Zeng Shumei, Luofang Lv, Liqing Yue, Hu Jian zhong, Huang Weihong

    出版事項 2025-07-01

    Objective To develop and validate an interpretable machine learning (ML)-based frailty risk prediction model that combines real-time health data with validated scale assessments for enhanced decision-making and targeted health management in integrated medical and older adult care institutions (IMOAC...

    詳細記述

    “…Hyperparameter optimisation was performed using stratified fivefold cross-validation with grid search, incorporating class-weighted loss functions to address class imbalance and model-specific regularisation techniques to maximise performance while preventing overfitting. …”
    全文の入手
    論文