Mismatch-Tolerant SNNs

This repository contains the code used in "Genetic Motifs as a Blueprint for Mismatch-Tolerant Neuromorphic Computing".

---

Note

The code provided in this repository is experimental and intended as a resource for readers interested in studying the methodology proposed. Please note that some of the terminology used in the code may differ from that used in the paper.

---

Repository Content

Below is a brief description of the main files included in the repository:

• models.py: Implementation of the proposed mismatch-tolerant architecture and of the spiking MLP used as baseline.
• train.py: Script for training the models.

Citation

@misc{boccato2024geneticmotifsblueprintmismatchtolerant,
      title={Genetic Motifs as a Blueprint for Mismatch-Tolerant Neuromorphic Computing}, 
      author={Tommaso Boccato and Dmitrii Zendrikov and Nicola Toschi and Giacomo Indiveri},
      year={2024},
      eprint={2410.19403},
      archivePrefix={arXiv},
      primaryClass={cs.NE},
      url={https://arxiv.org/abs/2410.19403}, 
}