Results 201 to 210 of about 1,925,652 (255)

Q‐LEAP: Millisecond Hyperdimensional Optimization for Full‐Spectrum Optical Metamaterials

Advanced Science, EarlyView.
Q‐LEAP integrates physics‐informed residual machine learning with factorization‐machine‐encoded quantum annealing to design full‐spectrum optical metamaterials. It explores a 2108 design space and, in a single 2.56 ms annealing step, reaches 85.83% of the theoretical FoM limit, enabling selective 5‐8 µm emission with 3–5 and 8–14 µm suppression and ∼40×
Zikang Guo, Yunwen Wu, Yuhui Guo, Shenghong Ju   +3 more
wiley   +1 more source

Transformer‐Based PET/CT Fusion Enables Preoperative Prediction and Prognostic Stratification of Spread Through Air Spaces in Lung Adenocarcinoma

Advanced Science, EarlyView.
An eight‐stream 2.5D Transformer integrates complementary PET metabolic and CT morphologic information to estimate spread‐through‐air‐spaces risk before surgery in lung adenocarcinoma. The model generalizes across independent centers, while its continuous risk score independently stratifies progression‐free survival, supporting further prospective ...
Xin‐Yu Zhu, Meng Meng, Zhen‐Zhen Wang, Deng‐Lu Lu, Ya‐Min Wei, Xu‐Chang Mi, Xing‐Yu Mu, Wei Fu   +7 more
wiley   +1 more source

An Integrated Artificial Intelligence Framework for Nanomedicine Engineering

Advanced Science, EarlyView.
This review outlines a closed‐loop AI ecosystem for nanomedicine engineering. By synergizing multimodal data harmonization, objective‐driven AI architectures, and automated high‐throughput validation, this “Design‐Make‐Test‐Analyze” (DMTA) workflow transforms traditional heuristic formulation into a predictive, self‐optimizing paradigm, accelerating ...
Meichen Dong, Qin Luo, Xinyi Ding, Chaoqi Lu, Ru Zhang, Qingqing Tan, Zhiqing Pang, Defang Ouyang, Dongsheng Li, Jiaxuan Xia, Jianxin Wang   +10 more
wiley   +1 more source

Accelerating hyperparameter optimization with a secretary

Neurocomputing
Víctor Muñoz, Carmen Ballester, Dorin Sabin Copaci, Luis Moreno 0001, Dolores Blanco   +4 more
openaire   +2 more sources

CausalCellInfer: Prioritizing Cell‐Type‐Specific Disease Associations From Genetically Predicted Transcriptomes in Biobank Cohorts

Advanced Science, EarlyView.
Bulk and genotype‐imputed transcriptomes obscure the cellular drivers of complex disease. CausalCellInfer resolves them by integrating invariant marker selection, deep learning deconvolution, and regularized matrix completion. Applied to ∼500 000 UK Biobank participants across 29 traits, it uncovers hallmark cell‐type proportion shifts, including ...
Liangying Yin, Yujia Shi, Ruoyu Zhang, Yong Xiang, Jinghong Qiu, Pak‐Chung Sham, Hon‐Cheong So   +6 more
wiley   +1 more source

Accelerated Discovery of Earth‐Abundant Oxygen Evolution Catalysts Through Surface‐Informed Active Learning

Advanced Science, EarlyView.
A surface‐informed active learning framework integrated with density functional theory enables resource efficient screening of the NiCoFe ternary composition space. Experimental validation identifies Ni55Co10Fe35 as the optimal composition, exhibiting favorable oxygen evolution reaction kinetics.
Hayun Jeon, Minhee Park, Won Bo Lee, Seyoung Park, YongJoo Kim, Sungwook Mhin   +5 more
wiley   +1 more source

Probe‐Free, Primer‐Programmed Amplicon Encoding Enables Nanopore‐Based Multiplex Detection With Deep Learning Decoding

Advanced Science, EarlyView.
Primer‐programmed amplicon encoding enables reporter‐probe‐free multiplex nucleic acid identification using glass nanopore sensing and deep neural network decoding. ABSTRACT Single‐reaction multiplex nucleic acid testing (NATs) must both co‐amplify multiple targets and distinguish the resulting amplicons within the product mixture.
Md. Ahasan Ahamed, Muhammad Asad Ullah Khalid, Wahid Uz Zaman, Anthony J. Politza, Weihua Guan   +4 more
wiley   +1 more source

Adaptive Hyperparameter Optimization

This paper introduces an adaptive hyperparameter optimization algorithm designed to enhance model generalization. Traditional hyperparameter optimization methods often employ static strategies, failing to adapt to the intricacies of individual models.
openaire   +1 more source

Steric‐Pressure‐Driven Topological Reconstruction for High‐Performance Zintl Phase Thermoelectrics with Machine‐Learning Validation

Advanced Science, EarlyView.
Synergistic Sr and Ag co‐doping drives steric‐pressure‐induced reconstruction from a 2D layered structure to a cross‐linked 3D framework in the Ca9‐xSrxZn4+y‐zAgzSb9 system, enhancing carrier mobility while maintaining a large Seebeck coefficient and doubling ZT to 0.73 at 873 K, with hybrid transfer machine learning further capturing the compositional
Siyeon Kim, Aziz Ahmed, Juwon Lee, Jihyun Lee, Ji Hee Pi, Kyu Hyoung Lee, Kang Min Ok, Tae‐Soo You   +7 more
wiley   +1 more source

Atomic‐Scale Mechanisms of Anisotropic Thermal Decomposition in GeSn Alloys With Stepwise Pinning

Advanced Electronic Materials, EarlyView.
Combining in situ TEM with DFT calculations, this work elucidates atomic‐scale anisotropic thermal decomposition in GeSn films, identifying laminar receding in defect‐free regions and stepwise pinning at stacking faults. Driven by crystallographic anisotropy and defect‐mediated energetic penalties, these findings establish a physical framework for ...
YiXin Wang, Qi Liu, XiYan Chang, Chi Xu, Tao Wang, XinZheng Li, ChunLai Xue   +6 more
wiley   +1 more source