Reproduce “Sector-Ring Linear Reservoirs: Sector-Wise Decay-Timescale Allocation With State Reuse Across Candidate Settings” 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
Yudai Ebato, Sou Nobukawa, Yusuke Sakemi, Takashi Kanamaru, Nina Sviridova, Kazuyuki Aihara
SRLR assigns groups of modes different decay timescales and reuses their state trajectories to reduce reservoir-search work. Across two chaotic systems and eight forecasting and reconstruction cases, it was compared with a common-timescale reference and nonlinear reservoir baselines. In Lorenz x-to-x forecasting at horizon 12, mean test NRMSE was 0.0209 versus 0.0283; SRLR’s mean error was lower than the reference in all eight cases, but not every baseline.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.