Code and Data for "Renormalization-Invariant Complexity: A Multiscale Criterion for Distinguishing Emergent Structure from Noise"

Reproducibility package for the paper "Renormalization-Invariant Complexity (RIC): A Multiscale Information-Theoretic Criterion for Distinguishing Emergent Structure from Noise" by K. Vasilchenko (Holon Institute of Technology).
The package contains a complete Jupyter notebook (Python) reproducing all results, figures, and statistical analyses reported in the manuscript. All simulated experiments (Kuramoto model, 2D Ising model, Hopfield network, scale-selective synthetic signal) are self-contained with fixed random seeds (base = 42).
For the experimental EEG validation, the paper uses three sets (Z, O, S) from the publicly available Bonn University Epilepsy Dataset (Andrzejak et al., 2001; https://repositori.upf.edu/handle/10230/42894). These three sets are included here as a convenience copy so that the notebook runs without any external downloads. The full Bonn dataset comprises five sets (Z, O, N, F, S); only the subset used in the analysis is provided.
Contents:
RIC_Framework_v4_4.ipynb — full analysis code (Kuramoto model, 2D Ising model, Hopfield network, scale-selective synthetic experiment, Bonn EEG analysis, binarization robustness, MSE/MPLZC comparison)data/Z.zip, data/O.zip, data/S.zip — Bonn University Epilepsy Dataset (Andrzejak et al., 2001): Sets Z (eyes open), O (eyes closed), S (seizure), 100 segments × 4097 samples at 173.61 Hz eachAll simulated experiments are self-contained with fixed random seeds (base = 42). No external downloads or API keys are required. Dependencies: NumPy, SciPy, Matplotlib, pandas, tqdm.