E-prop plasticity examples

Eligibility propagation (e-prop) [1] is a three-factor learning rule for spiking neural networks that approximates the performance of backpropagation through time (BPTT). A complete e-prop model comprises a recurrent neuron model, a readout neuron model, a synapse model, and a learning signal connection. Two such models are provided: the original formulation by Bellec et al. (2020) [1] and an extended variant with additional biological features [2]. The e-prop models are related as follows:

We provide tutorials to reproduce the supervised regression task for generating temporal patterns and the supervised classification task from the original TensorFlow implementation. In addition, we provide two tutorials on supervised regression for generating two-dimensional temporal patterns and on supervised classification of neuromorphic MNIST [3].

References

[1] (1,2)

Guillaume Bellec, Franz Scherr, Anand Subramoney, Elias Hajek, Darjan Salaj, Robert Legenstein, and Wolfgang Maass. A solution to the learning dilemma for recurrent networks of spiking neurons. Nature Communications, 11:3625, 2020. doi:10.1038/s41467-020-17236-y.

[2]

Agnes Korcsak-Gorzo, Jesús A. Espinoza Valverde, Jonas Stapmanns, Hans Ekkehard Plesser, David Dahmen, Matthias Bolten, Sacha J. van Albada, and Markus Diesmann. Event-driven eligibility propagation in large sparse networks: efficiency shaped by biological realism. arXiv, 2025. doi:10.48550/arXiv.2511.21674.

[3]

Garrick Orchard, Ajinkya Jayawant, Gregory K. Cohen, and Nitish Thakor. Converting static image datasets to spiking neuromorphic datasets using saccades. Frontiers in Neuroscience, 9:437, 2015. doi:10.3389/fnins.2015.00437.

Tutorial on learning to generate sine waves with e-prop after Bellec et al. (2020)

Tutorial on learning to generate sine waves with e-prop after Bellec et al. (2020)

Tutorial on learning to generate sine waves with e-prop

Tutorial on learning to generate sine waves with e-prop

Tutorial on learning to generate a lemniscate with e-prop after Bellec et al. (2020)

Tutorial on learning to generate a lemniscate with e-prop after Bellec et al. (2020)

Tutorial on learning to generate handwritten text with e-prop after Bellec et al. (2020)

Tutorial on learning to generate handwritten text with e-prop after Bellec et al. (2020)

Tutorial on learning to accumulate evidence with e-prop after Bellec et al. (2020)

Tutorial on learning to accumulate evidence with e-prop after Bellec et al. (2020)

Tutorial on learning N-MNIST classification with e-prop

Tutorial on learning N-MNIST classification with e-prop