Cognitive computing building block: A versatile and efficient digital neuron model for neurosynaptic cores

TLDR

IBM advances the DARPA SyNAPSE roadmap with a trilogy of innovations toward the TrueNorth cognitive computing system, drawing inspiration from brain function and efficiency. The authors develop a simple, digital, reconfigurable, versatile spiking neuron model that balances functional capability and implementation cost, enabling one-to-one hardware–simulation equivalence with only 1,272 ASIC gates. The model builds on a leaky integrate‑and‑fire neuron, adding configurable stochasticity, four leak modes, deterministic and stochastic thresholds, six reset modes, and a library of over 50 neuron behaviors for hierarchical composition. It supports a wide range of computational functions and neural codes and can qualitatively replicate 20 biologically relevant behaviors of a dynamical neuron model.

Abstract

Marching along the DARPA SyNAPSE roadmap, IBM unveils a trilogy of innovations towards the TrueNorth cognitive computing system inspired by the brain's function and efficiency. Judiciously balancing the dual objectives of functional capability and implementation/operational cost, we develop a simple, digital, reconfigurable, versatile spiking neuron model that supports one-to-one equivalence between hardware and simulation and is implementable using only 1272 ASIC gates. Starting with the classic leaky integrate-and-fire neuron, we add: (a) configurable and reproducible stochasticity to the input, the state, and the output; (b) four leak modes that bias the internal state dynamics; (c) deterministic and stochastic thresholds; and (d) six reset modes for rich finite-state behavior. The model supports a wide variety of computational functions and neural codes. We capture 50+ neuron behaviors in a library for hierarchical composition of complex computations and behaviors. Although designed with cognitive algorithms and applications in mind, serendipitously, the neuron model can qualitatively replicate the 20 biologically-relevant behaviors of a dynamical neuron model.