Code for training a network on the Heidelberg spoken digits spiking database, and attacking the data with a sparse adversarial attack that fools the network.
The majority of the training code, with modifications, is by Manu Srinath Halvagal (https://zenkelab.org/team/) and Friedemann Zenke (https://fzenke.net), who released it under the Creative Commons Attribution 4.0 International License. See this file.
The sparsefool code, with some modifications, comes from the LTS4 lab at EPFL, who released it under the Apache 2.0 license. See this repo.
For more details on surrogate gradient learning, please see:
- Neftci, E.O., Mostafa, H., and Zenke, F. (2019). Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient based optimization to spiking neural networks. IEEE Signal Process Mag 36, 51–63. https://ieeexplore.ieee.org/document/8891809 and https://arxiv.org/abs/1901.09948
For more details on SparseFool see:
- Apostolos Modas, Seyed-Mohsen Moosavi-Dezfooli, Pascal Frossard (2019). SparseFool: a few pixels make a big difference. CVPR 2019. https://arxiv.org/abs/1811.02248.
For more details on the Heidelberg dataset see:
- Cramer, B., Stradmann, Y., Schemmel, J., and Zenke, F. (2020). The Heidelberg Spiking Data Sets for the Systematic Evaluation of Spiking Neural Networks. IEEE Transactions on Neural Networks and Learning Systems 1–14. https://ieeexplore.ieee.org/document/9311226