Reproduce “Physics-Informed Echo State Networks for Chaotic Systems Forecasting” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Nguyen Anh Khoa Doan, Wolfgang Polifke, Luca Magri
The authors develop a physics-informed Echo State Network (ESN) for forecasting chaotic systems after training. It predicts system states while penalizing residuals from known governing equations along its own rollout, without additional training data. In a Lorenz test with 200 reservoir units, the predictability horizon increased from 4 to approximately 5.5 Lyapunov times, relative to the tested conventional ESN.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.