Event-based learning of delays in SNNs

In a collaboration between Melika Payvand's EIS Lab and Mihai Petrovici's NeuroTMA group, we've been developing a fully event-driven training algorithm for learning delays and weights in SNNs. In the established Fast&Deep algorithm, spike times are treated as the quantity central for information propagation. From this perspective, learning of delays appears naturally and can profit from the efficiency of event-based formalisms over time-stepped algorithms, while maintaining mathematical exactness.

In this workshop we want to present our findings on the efficacy of delays, show how delay-learning appears natively in an event-based formalism, apply our method to datasets of other participants to test its applicability and hope to implement our work on available hardware.