Results 71 to 80 of about 3,592 (172)

Deep Learning–Based Extraction of Promising Material Groups and Common Features from High‐Dimensional Data: A Case of Optical Spectra of Inorganic Crystals

Advanced Intelligent Discovery, EarlyView.
We report a novel interpretation method for deep learning models based on feature extraction and clustering. Applying this method to an atomistic line graph neural network (ALIGNN) model trained on optical absorption spectra of 2,681 inorganic compounds obtained from first‐principles calculations, we successfully identify key factors underlying ...
Akira Takahashi, Yu Kumagai, Arata Takamatsu, Fumiyasu Oba   +3 more
wiley   +1 more source

RiemannInfer: improving transformer inference through Riemannian geometry. [PDF]

Sci Rep
Mao R, Zhang Z, Yang M, Xie H, Wei S, Hu C.   +5 more
europepmc   +1 more source

Lidar‐Based Object Tracking of Traffic Participants with Sensor Nodes in Existing Urban Infrastructure

Advanced Intelligent Systems, EarlyView.
This paper presents a lidar‐based sensor node design and a rule‐based state observer for edge‐based traffic participant tracking. Unlike other state‐of‐the‐art methods, this state observer enables real‐time, CPU‐only edge processing without relying on machine learning approaches.
Simon Schäfer, Bassam Alrifaee, Ehsan Hashemi   +2 more
wiley   +1 more source

Multi-order hyperbolic graph convolution and aggregated attention for social event detection. [PDF]

PLoS One
Liu Y, Tan TP, Liu Z, Li Y.
europepmc   +1 more source

Resource‐Aware Contrastive Scattering Meta‐Learning for Efficient Few‐Shot Acoustic Anomaly Detection

Advanced Intelligent Systems, EarlyView.
This paper introduces a resource‐aware Contrastive Scattering Meta‐Learning (CSML) framework for acoustic anomaly detection. By leveraging training‐free wavelet scattering and metric‐based meta‐learning, the model achieves competitive performance with only 50 K learnable parameters—a 98% reduction compared to state‐of‐the‐art frameworks—enabling ...
Rami Zewail, Bassem Mokhtar
wiley   +1 more source

Advances in Optimized and Safe Path Planning of Marine Autonomous Surface Vehicles: A Review. [PDF]

Sensors (Basel)
Kou L, Gao X.
europepmc   +1 more source

Pre‐Curved Everting Robots With Embedded Steering Intelligence Fabricated by CO2 Laser Welding

Advanced Intelligent Systems, EarlyView.
Design and experimental demonstration of a laser welded growing robot for anatomically guided navigation. The robot follows an aortic arch phantom entering the branchiocephalic branch through steering by design. The figure shows the physical phantom setup, CAD defined weld geometry and full robot eversion.
Brandon Saldarriaga, Panagiotis Kalozoumis, Panagiotis Vartholomeos, Carlo Seneci, Kawal Rhode, Christos Bergeles   +5 more
wiley   +1 more source

An Integrated and Robust Deep Learning Framework for Denoising and Analyzing Single‐Cell Spatial Transcriptomics

Advanced Intelligent Systems, EarlyView.
Single‐cell Spatial Transcriptomics Analysis and Denoising Engine is introduced as a unified deep learning framework that jointly performs denoising, clustering, and gene prioritization in spatial transcriptomics. By integrating linear and nonlinear representations within a dual‐channel architecture, it improves robustness and accuracy, uncovers ...
Yaxuan Cui, Yang Cui, Ruheng Wang, Xin Zeng, Zheyong Zhu, Ding Yi, Ruihua Guo, Tetsuya Sakurai, Kenta Nakai, Xiucai Ye, Leyi Wei, Yong Chen   +11 more
wiley   +1 more source

Diffusion-Shock PDEs for Deep Learning on Position-Orientation Space. [PDF]

J Math Imaging Vis
Sherry FM, Schaefer K, Duits R.
europepmc   +1 more source

Overcoming the Phase Discontinuity Problem in RIS Optimization: A Robust and Scalable Phase‐Aware Deep Regression Framework via Hybrid CNN–LSTM Architecture

Advanced Intelligent Systems, EarlyView.
The phase discontinuity problem—where the cyclic nature of phase angles causes catastrophic errors near the ±π boundary—is a fundamental obstacle in learning‐based reconfigurable intelligent surface (RIS) optimization. A phase‐aware hybrid CNN–LSTM framework resolves this by decomposing phase predictions into sine–cosine components, mapping circular ...
Seda Savaşçı Şen, Ali Çalhan, Murtaza Cicioğlu, Oğuzhan Demiryürek   +3 more
wiley   +1 more source