Results 61 to 70 of about 3,330 (201)

Unraveling the Spatiotemporal Dynamics and Nonlinear Drivers of Green Transition of Farmland Use in Major Grain‐Producing Areas: A Case Study of Jiangsu Province, China

Land Degradation &Development, EarlyView.
ABSTRACT Promoting the green transition of farmland use (GTFU) in major grain‐producing areas is essential for ensuring food security and advancing sustainable agricultural development. However, existing studies on GTFU have predominantly relied on static cross‐sectional analyses, with insufficient attention paid to its nonlinear driving mechanisms. To
Zhixian Sun, Yixuan Lu, Xinyue Zheng, Dongmei Zhang, Yian Xu, Guanxin Yao   +5 more
wiley   +1 more source

Optimization of Scene and Material Parameters for the Generation of Synthetic Training Datasets for Machine Learning-Based Object Segmentation

Computers
Synthetic training data is often essential for neural-network-based segmentation when real datasets are difficult or impossible to obtain. Conventional synthetic data generation relies on manually selecting scene and material parameters. This can lead to
Malte Nagel, Kolja Hedrich, Nils Melchert, Lennart Hinz, Eduard Reithmeier   +4 more
doaj   +1 more source

BayesianKAN: A reaction condition optimization framework integrating Kolmogorov‐Arnold network and Bayesian optimization

Smart Molecules, EarlyView.
A novel BayesianKAN framework integrates Kolmogorov‐Arnold networks with Bayesian optimization to efficiently navigate complex factor spaces, accelerating the discovery of optimal reaction conditions for chemical synthesis. Abstract Efficient optimization of chemical reaction conditions is crucial for enhancing reaction yield and selectivity, yet ...
Juntao Wang, Yujing Zhao, Peiyu Yi, Liyuan Duan, Qingwei Meng, Qilei Liu   +5 more
wiley   +1 more source

Decreasing the Environmental Impact of the Electric Steelmaking Route Through Advanced Modelling Techniques

steel research international, EarlyView.
Ensemble models are adopted to estimate the sterile content of scraps arriving to the scrap yard. Feed‐forward neural networks are exploited to estimate steel composition and temperature after Ladle furnace. The models are validated on data from two steelworks very satisfactory results and are inherently transferable to other steelworks, as they are ...
Valentina Colla, Antonella Zaccara, Stefano Dettori, Laura Laid, Ismael Matino, Silvia Cateni, Teresa Annunziata Branca, Lorenzo Vannini   +7 more
wiley   +1 more source

Fine‐tuning ab initio XANES spectra calculations using the Bayesian optimization algorithm

Journal of Synchrotron Radiation, EarlyView.
A Bayesian optimization technique is used to tune the FEFF and FDMNES packages and to improve matching between theoretical and experimental spectra. The tests were performed on monometallic Ni, Fe and Pd K‐edge XANES spectra using several different spectrum similarity metrics.Theoretical modeling of X‐ray absorption near‐edge structure (XANES) spectra ...
Andrey A. Sapronov, Dmitry E. Doronkin, Jan-Dierk Grunwaldt   +2 more
wiley   +1 more source

Hybrid Machine Learning Approach for Nutrient Deficiency Detection in Lettuce

TIERS Information Technology Journal
Early detection of nutrient deficiencies in lettuce is essential for precision agriculture. However, this task remains challenging due to limited data availability and class imbalance, which reduce model sensitivity toward minority classes and hinder ...
Zuriati Zuriati, Dewi Kania Widyawati, Oki Arifin, Kurniawan Saputra, Sriyanto Sriyanto, Asmala Ahmad   +5 more
doaj   +1 more source

Enhanced Modelling Performance with Boosting Ensemble Meta-Learning and Optuna Optimization

SN Computer Science
AbstractImproving modeling performance on imbalanced multi-class classification problems has continued to attract attention from researchers considering the critical and significant role such models should play in mitigating the prevalent problem. Ensemble Learning (EL) techniques are among the key methods utilized by researchers as they are known for ...
Tertsegha J. Anande, Sami Alsaadi, Mark S. Leeson   +2 more
openaire   +1 more source

Multimodal Video Summarization Using Vision‐Language Embeddings and Hierarchical Temporal Modeling

Applied AI Letters, Volume 7, Issue 3, October 2026.
Combining BLIP‐2 image captions with CLIP vision–language embeddings gives video frames semantic meaning that pixels alone cannot convey. Processed by a multi‐scale temporal U‐Net and hierarchical shot‐aware transformer, these multimodal features achieve a state‐of‐the‐art 59.27% F1‐score on SumMe using only 28.27M parameters.
Saadman Sakib, Kaushik Deb
wiley   +1 more source

Optuna Tuning Results PPO Reinforcement Learning Hyperparameters Performance

Systematic hyperparameter tuning using Optuna was expected to improve PPO model performance in a multi-microgrid environment. We hypothesized that optimizing hyperparameters like learning rate and network architecture would enhance model performance ...
Messlem, A (via Mendeley Data)
core   +1 more source

Optuna-Optimized Machine Learning for Traffic Accident Severity Prediction in Jordan under Class Imbalance Conditions

Communications
A machine learning framework for predicting the traffic accident severity under class imbalance conditions is predented in Amman, Jordan. The methodology began with a preprocessing pipeline consisting of IQR-based (Interquartile range) outlier removal ...
Maen Qaseem Ghadi
doaj   +1 more source