MetaSeq generates acoustic metamaterial structures from target broadband responses via a sequence representation and physics-guided RL, cutting response error 45% below the best of five baselines in COMSOL evaluations.
Gelu activation function in deep learning: a com- prehensive mathematical analysis and performance.arXiv preprint arXiv:2305.12073, 2023
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
NFMR-Net refines coarse LoS-based range and angle estimates from ZC-pilot channel matrices and 2D MUSIC to outperform standard 2D MUSIC in wideband near-field multi-user localization.
An encoder-decoder neural network trained on boundary element method data designs and compares layered cloaks for 2D Helmholtz scattering, showing object-fitted layers reduce scattering more than circular ones for circular, star, and kite objects.
citing papers explorer
-
Physics-Guided Sequence-Based Generative Framework for Acoustic Metamaterial Inverse Design
MetaSeq generates acoustic metamaterial structures from target broadband responses via a sequence representation and physics-guided RL, cutting response error 45% below the best of five baselines in COMSOL evaluations.
-
Near-field Wideband Multi-User Localization using NFMR-Net
NFMR-Net refines coarse LoS-based range and angle estimates from ZC-pilot channel matrices and 2D MUSIC to outperform standard 2D MUSIC in wideband near-field multi-user localization.
-
Solving forward and inverse wave scattering via boundary integral equations and deep learning. Applications to cloaking design
An encoder-decoder neural network trained on boundary element method data designs and compares layered cloaks for 2D Helmholtz scattering, showing object-fitted layers reduce scattering more than circular ones for circular, star, and kite objects.