Julia implementation of a scaled projection neural network for quasi-variational inequalities with state-dependent constraint set S(x) = m(x) + S and fixed symmetric positive-definite matrix M. Integrates the continuous-time dynamics dx/dt = lambda * [P_{S(x),M^{-1}}(x - alpha * M * F(x)) - x] (projection in the M^{-1} metric) by adaptive ODE integration (OrdinaryDiffEq.jl) and by a discrete Krasnosel'skii-Mann iteration; sparse obstacle-type projections are computed by a KKT-verified primal-dual active-set method.
Abstract: 5G is rolled out and next generation 6G networks are also being developed, ultra-low latency (URLL) communication as a standard is critical in supporting the plethora of applications, spanning autonomous vehicles, immersive extended reality experience, etc. However, tra…
We present a Graph Neural Network (GNN) framework for the classification of finite groups according to their solvability. Using undirected Cayley graph representations, the proposed framework learns to distinguish solvable and non-solvable groups directly from structural graph in…
The increasing reliance on third-party packages from repositories such as Python Package Index (PyPI) and Node Package Manager (NPM) has introduced critical vulnerabilities in software supply chains. Traditional security approaches, including signature-based detection and trust e…
This deposit contains the field data, synthetic training datasets, trained network weights and analysis code supporting the article "Physics-informed neural network inversion of electrical resistivity tomography data: amortized optimization with field validation in the Moroccan M…
Announcement: Revised Edition of the Technical Note Published on Zenodo A revised edition of the technical note “A Construction Method for Neural Networks Without Activation Functions” has been published on Zenodo. This updated version includes a newly added Appendix, which provi…
We present a comprehensive automated solution for 3D seismic fault detection and interpretation that combines deep learning with advanced geometric post-processing. The method integrates a 3D U-Net neural network trained on synthetic data with normalized distance function targets…