3. Event-based sensing

We have seen how neurons can compute and how we can optimize their spatial and temporal components. We also saw that interneural communication happens with discrete events, so-called “spikes”. Now, we will see how sensors can provide us with spiking data and how we can work with that using our newfound knowledge about neurons and computation.

Lecture content

1. Working with AER

2. Temporal dynamics

3. Project 3: Predicting digits with time surfaces

Literature

• Training Spiking Neural Networks Using Lessons From Deep Learning

• SuperSpike: Supervised Learning in Multilayer Spiking Neural Networks

• Neuromorphic Engineering: In Memory of Misha Mahowald

• Translation and Scale Invariance for Event-Based Object tracking