New SPCNNs Architecture for Approximating Polyconvex Energies

Researchers have developed a new type of neural network architecture called Symmetric Polyconvex Neural Networks (SPCNNs) to address the challenges of approximating symmetric polyconvex energies. These networks combine a convexity property with the monotonicity required for representing minors, overcoming limitations of previous ReLU-based networks. The SPCNNs are proven to universally approximate symmetric polyconvex energies without assuming isotropy, and their accuracy is demonstrated through numerical examples involving stress-based regression and the approximation of a double-well potential's symmetric polyconvex envelope. AI

IMPACT Introduces a novel neural network architecture for complex energy approximations, potentially advancing computational mechanics and material science.

RANK_REASON The cluster contains a research paper detailing a new neural network architecture. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• DagsHub
• Gotit.pub
• Hugging Face
• Litmaps
• ScienceCast
• scite Smart Citations
• SPCNNs
• Symmetric Polyconvex Neural Networks

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →