Researchers have developed parameter-conditioned Fourier Neural Operators (FNOs) to create fast, differentiable surrogate models for the FitzHugh-Nagumo (FHN) system. These models can accurately simulate neuronal voltage dynamics, including excitable and oscillatory regimes, with significantly reduced computational cost compared to traditional solvers. The FNOs achieve sub-0.1% relative L2 error in the oscillatory regime and accurately reproduce key characteristics like firing thresholds and conduction velocities in the excitable regime, demonstrating strong generalization and extrapolation capabilities. AI
IMPACT Accelerates scientific discovery by enabling faster simulation of complex biological systems.
RANK_REASON The cluster contains an academic paper detailing a new method for modeling complex dynamics using neural networks. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →