LiL-Q applies quasilinearization to nonlinear PDEs and solves each resulting linear problem by convex least-squares collocation on Linear-in-Learnables trial spaces, achieving fast convergence and high accuracy on multiple benchmarks.
Extreme learning machine: Theory and applications , year =
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
verdicts
UNVERDICTED 8representative citing papers
A free-space PELM achieves 96.56% on MNIST, 95.67% on spoken digit spectrograms, 100% on mushroom classification, and 0.0699 NRMSE on abalone regression using the same optical setup, claimed as the first multimodal free-space PELM.
FLASH-MAX embeds exact Maxwell solutions as neurons in a neural network to reconstruct homogeneous EM fields from sparse data with guaranteed zero PDE residual and proven universal approximation on arbitrary domains.
HTC predicts PNN classification loss via a power law, with experimental and simulated data from distinct physical systems collapsing onto task-specific curves.
A hybrid FEM and ELM framework for parameter-dependent PDEs derives existence, uniqueness, regularity, and error estimates for inverse problems in photoacoustic tomography.
The authors introduce dRVFL and edRVFL frameworks that stack RVFL layers with fixed random weights and closed-form outputs, reporting superior benchmark performance when combined with sparse-pretrained RVFL.
R-HessELM with inclined entropy features predicts CHF from ECG signals with 98.49% accuracy.
Replacing SVD in ELM with LU, Hessenberg, Schur, Gram-Schmidt or Householder decompositions speeds training on large EEG BCI data, with Hessenberg preferred for pace and Householder for accuracy.
citing papers explorer
-
A Convex Quasilinearization Method for Solving Nonlinear PDEs with Physics-Informed Neural Networks
LiL-Q applies quasilinearization to nonlinear PDEs and solves each resulting linear problem by convex least-squares collocation on Linear-in-Learnables trial spaces, achieving fast convergence and high accuracy on multiple benchmarks.
-
Multimodal Optical Feature Extraction with a Free-Space Photonic Extreme Learning Machine
A free-space PELM achieves 96.56% on MNIST, 95.67% on spoken digit spectrograms, 100% on mushroom classification, and 0.0699 NRMSE on abalone regression using the same optical setup, claimed as the first multimodal free-space PELM.
-
Fast Reconstruction of Exact Maxwell Dynamics from Sparse Data
FLASH-MAX embeds exact Maxwell solutions as neurons in a neural network to reconstruct homogeneous EM fields from sparse data with guaranteed zero PDE residual and proven universal approximation on arbitrary domains.
-
Power law scaling for classification accuracy in physical neural networks
HTC predicts PNN classification loss via a power law, with experimental and simulated data from distinct physical systems collapsing onto task-specific curves.
-
Solving Inverse Parametrized Problems via Finite Elements and Extreme Learning Networks
A hybrid FEM and ELM framework for parameter-dependent PDEs derives existence, uniqueness, regularity, and error estimates for inverse problems in photoacoustic tomography.
-
Random Vector Functional Link Neural Network based Ensemble Deep Learning
The authors introduce dRVFL and edRVFL frameworks that stack RVFL layers with fixed random weights and closed-form outputs, reporting superior benchmark performance when combined with sparse-pretrained RVFL.
-
Regularized HessELM and Inclined Entropy Measurement for Congestive Heart Failure Prediction
R-HessELM with inclined entropy features predicts CHF from ECG signals with 98.49% accuracy.
-
On improving learning capability of ELM and an application to brain-computer interface
Replacing SVD in ELM with LU, Hessenberg, Schur, Gram-Schmidt or Householder decompositions speeds training on large EEG BCI data, with Hessenberg preferred for pace and Householder for accuracy.