1/4 Surrogate gradients are a great tool for training spiking neural networks in comp neuro and neuromorphic engineering, but what is a good initialization? In our new preprint co-led by Julian Rossbroich and Julia Gygax, we lay out practical strategies: arxiv.org/abs/2206.10226

Are you at the #BernsteinConference? If you are interested in understanding surrogate gradients for spiking neural network training stop by my poster C13 on Wed, Sep. 27 at 6pm

“The combination of Hebbian and predictive plasticity learns invariant object representations in deep sensory networks” is now published in Nat Neurosci. We show how local plasticity mechanisms can learn invariant representations through prediction. nature.com/articles/s4159…

1/6 We updated “Implicit variance regularization in non-contrastive SSL,” by Manu Halvagal and Axel Laborieux with new results for NeurIPS camera-ready: arxiv.org/abs/2212.04858 neurips.cc/virtual/2023/p…

Very excited to share this new paper, in which we propose a possible role for (dis-)inhibitory microcircuits in coordinating plasticity. If you want to discuss, come find me NeurIPS Conference or drop us a message.

1/6 Surrogate gradients (SGs) are empirically successful at training spiking neural networks (SNNs). But why do they work so well, and what is their theoretical basis? In our new preprint led by Julia Gygax, we provide the answers: arxiv.org/abs/2404.14964

We're hiring! Come build models of how the brain learns and simulates a world model. We have several openings at PhD and postdoc levels, including a collab with Georg Keller lab on designing regulatory elements to target distinct neuronal cell types. zenkelab.org/jobs

Thank you so much for choosing our work for the Ruth Chiquet Prize, it's a big honour. And thanks to Tengjun Liu and Julian Rossbroich for the wonderful collaboration.

1/3 Brain-machine interfaces could improve the lives of locked-in patients. Yet, to minimize the risk of infection, we need fully implanted systems that are accurate and energy efficient. We show that spiking neural networks are ideal: arxiv.org/abs/2409.01762 (accepted at BioCAS)

The #SNUFA24 final program is out and the event is next Tue-Wed! If you love spiking neural networks, click on the link below to check it out and register (free). snufa.net/2024/