Tag: Neural Networks

Researchers have demonstrated that machine learning algorithms and symbolic regression can accurately predict the ultimate moment capacity and failure modes ...

Researchers have demonstrated that integrating noise-aware loss correction directly into collaborative filtering training outperforms traditional data cleansing when handling corrupted ...

Researchers report that replacing Adam's squared-gradient second moment with the mean absolute deviation yields smoother training, fewer oscillations and faster ...

Researchers report that a transfer learning framework called TL-CDVD reached an average accuracy of 88.8 percent and an F1-score of ...

An editorial in Frontiers in Computational Neuroscience gathers eight studies arguing that neural systems operate near a critical state between ...

Three tiny neural networks, each under 50,000 parameters, cut prediction error by more than half and reached up to 95 ...

The new OPCNet neural network fits highly oscillatory data by learning the dynamics of a linear ODE and fitting the ...

Simulations show that short, distance-dependent excitatory connections amplify shared inputs, drive burst propagation, and generate modular functional organization without built-in ...

Researchers have developed DynamicFusionNet, a decoder that adaptively fuses features across layers, scales and views to improve medical image segmentation ...

Virginia Tech researchers combined hammer-induced vibrations, laser measurements and a compact neural network to reliably estimate how deeply rocks are ...

A comprehensive review details how deep neural networks are automating every stage of cryo-electron microscopy, from particle picking to high-resolution ...

A new study reports that hybrid broad learning systems optimized by parameter-free metaheuristics predict concrete compressive strength with MAPE below ...

A three-stage physics-guided tomography framework reconstructs flame temperature and concentration with low errors from only 40 laser paths and two ...

Researchers in Stuttgart have introduced a machine-learning framework that identifies stable, passive port-Hamiltonian models directly from data, validated on systems ...

Researchers have developed CLeaR, a continual learning framework that lets neural networks handle regression tasks such as wind power forecasting ...

A sensitivity analysis using MLP and RBF neural networks on synthetic cooling data finds that temperature difference, airflow, and mass ...

A systematic review of mortgage-lending fairness research finds decision trees offer the most interpretable bias detection while neural networks and ...

Researchers propose SteRGAN, a GAN-based image steganography framework that achieves PSNR values above 41 dB on ImageNet and Microsoft COCO.

A single-cell projectome map of mouse cortical neurons reveals parallel, segregated brainstem projections that inspired an AI architecture with 21 ...

A simulation study published in Discover Computing reports that a hybrid framework combining a sorting-inspired data prioritization score with neural ...