This work has received funding from the EC Horizon 2020 Framework Programme under grant agreements 785907 (HBP SGA2) and 945539 (HBP SGA3), the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) under Germany's Excellence Strategy EXC 2181/1-390900948 (the Heidelberg STRUCTURES Excellence Cluster), the German Federal Ministry of Education and Research under grant number 16ES1127 as part of the Pilotinnovationswettbewerb Energieeffizientes KI-System, the Helmholtz Association Initiative and Networking Fund [Advanced Computing Architectures (ACA)] under Project SO-092, as well as from the Manfred Stärk Foundation, and the Lautenschläger-Forschungspreis 2018 for Karlheinz Meier.

jaxsnn is an event-based approach to machine-learning-inspired training and simulation of SNNs, including support for neuromorphic backends (BrainScaleS-2). We build upon jax, a Python library providing autograd and XLA functionality for high-performance machine learning research.