New LRX-PINN model improves accuracy for convection-dominated problems

Researchers have developed a new type of physics-informed neural network called LRX-PINN, designed to handle convection-dominated problems. This network utilizes integrated Cauchy activations to effectively model the thin layers and sharp transition profiles characteristic of such problems. LRX-PINN demonstrates improved accuracy and parameter efficiency compared to existing methods like PIKAN and Fourier-feature PINNs, and can be integrated into existing frameworks like hp-VPINN for further performance gains. AI

IMPACT This new neural network architecture offers a more efficient and accurate approach for solving complex fluid dynamics and similar problems.

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

Read on arXiv cs.LG →

• Cauchy activations
• Fourier-feature PINNs
• hp-VPINN
• LRX-PINN
• PIKAN
• Xnet

• paper
• other

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