Bio-Adaptive Quantum Error Correction: Immune-Inspired Priors Enable 22–65% Overhead Reduction in Surface-Code Decoding

This dataset and code repository accompanies the publication introducing BA-QEC, the first quantum-error-correction decoder explicitly inspired by biological immune-system architecture. BA-QEC integrates a Bayesian prior derived from human TCRβ CDR3 length distributions and an adaptive clonal-expansion memory mechanism to improve decoding performance in topological quantum codes. Simulations of a distance-7 rotated surface code demonstrate 22% threshold improvement from the biological prior alone, and up to 61% enhancement when combined with clonal memory under temporally correlated (1/f-type) noise. All code is open-source (MIT license) and fully reproducible in <10 minutes on Google Colab.

The repository includes:

Python notebooks for Stim-based and PyMatching-based simulations,

Clonal-expansion cache implementation,

Scripts for reproducing figures and pseudothreshold plots,

Documentation on integrating the biological prior into MWPM decoding.

This work establishes a novel link between adaptive immunity and quantum error correction, offering a new paradigm for biologically inspired, efficient, and adaptive decoders.