Covariant spatio-temporal receptive fields for spiking neural networks

Abstract

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.

Code available at https://github.com/jegp/nrf

References in corpus (13)

• Event-based Vision: A Survey
• HFirst: A Temporal Approach to Object Recognition
• Neuromorphic Intermediate Representation: A Unified Instruction Set for Interoperable Brain-Inspired Computing
• Time-causal and time-recursive spatio-temporal receptive fields
• Normative theory of visual receptive fields
• Scale-covariant and scale-invariant Gaussian derivative networks
• Discrete approximations of Gaussian smoothing and Gaussian derivatives
• Dynamic texture recognition using time-causal and time-recursive spatio-temporal receptive fields
• A time-causal and time-recursive scale-covariant scale-space representation of temporal signals and past time
• Spiking Neural Network Nonlinear Demapping on Neuromorphic Hardware for IM/DD Optical Communication
• Temporal scale selection in time-causal scale space
• Exploring the ability of CNNs to generalise to previously unseen scales over wide scale ranges
• Covariance properties under natural image transformations for the generalized Gaussian derivative model for visual receptive fields