Covariant spatio-temporal receptive fields for spiking neural networks
This work addresses the problem of inefficient neuromorphic system design for researchers in computational neuroscience and event-based vision, though it is incremental as it builds on existing scale-space theory and computational neuroscience efforts.
The paper tackled the lack of theoretical guidance for efficient neuromorphic implementations by developing a principled computational model using spatio-temporal receptive fields, and demonstrated that using these as a prior improves training of spiking networks in an event-based vision task.
Biological nervous systems constitute important sources of inspiration towards computers that are faster, cheaper, and more energy efficient. Neuromorphic disciplines view the brain as a coevolved system, simultaneously optimizing the hardware and the algorithms running on it. There are clear efficiency gains when bringing the computations into a physical substrate, but we presently lack theories to guide efficient implementations. Here, we present a principled computational model for neuromorphic systems in terms of spatio-temporal receptive fields, based on affine Gaussian kernels over space and leaky-integrator and leaky integrate-and-fire models over time. Our theory is provably covariant to spatial affine and temporal scaling transformations, and with close similarities to the visual processing in mammalian brains. We use these spatio-temporal receptive fields as a prior in an event-based vision task, and show that this improves the training of spiking networks, which otherwise is known as problematic for event-based vision. This work combines efforts within scale-space theory and computational neuroscience to identify theoretically well-founded ways to process spatio-temporal signals in neuromorphic systems. Our contributions are immediately relevant for signal processing and event-based vision, and can be extended to other processing tasks over space and time, such as memory and control.