AI discovers and rigorously certifies new water wave solutions

Researchers have developed a novel computational framework that merges machine learning with rigorous computer-assisted numerical certification and formal verification. This approach uses physics-informed neural networks to discover and refine solutions for nonlinear partial differential equations, specifically demonstrating its capability with periodic gravity-capillary traveling water waves. The system achieves near-machine-precision residuals, recovering known wave profiles like Stokes and Crapper waves, and identifying new candidate solutions. The accuracy of these neural network findings is independently validated through interval arithmetic for rigorous certification and further formalized using the Lean theorem prover, ensuring auditable mathematical guarantees. AI

IMPACT This research demonstrates a novel pipeline for AI-driven scientific discovery and rigorous mathematical verification, potentially accelerating breakthroughs in complex physical systems.

RANK_REASON Academic paper detailing a new methodology for scientific discovery and verification. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• Crapper waves
• DagsHub
• Gotit.pub
• Hugging Face
• Lean
• ScienceCast
• Wilton ripples

• paper
• other

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