An Accelerated Analog Neuromorphic Hardware System Emulating NMDA- and Calcium-Based Non-Linear Dendrites

TL;DR

This paper introduces an advanced neuromorphic hardware system that emulates complex dendritic behaviors and spike types of cortical neurons, enabling more biologically realistic neural computations in hardware.

Contribution

It extends the BrainScaleS neuromorphic platform with multi-compartment and non-linear dendrite support using a 65nm ASIC, allowing detailed emulation of neuron dynamics.

Findings

Successfully emulates NMDA and calcium spikes

Supports coincidence detection within single neurons

Demonstrates functionality through transistor-level simulations

Abstract

This paper presents an extension of the BrainScaleS accelerated analog neuromorphic hardware model. The scalable neuromorphic architecture is extended by the support for multi-compartment models and non-linear dendrites. These features are part of a \SI{65}{\nano\meter} prototype ASIC. It allows to emulate different spike types observed in cortical pyramidal neurons: NMDA plateau potentials, calcium and sodium spikes. By replicating some of the structures of these cells, they can be configured to perform coincidence detection within a single neuron. Built-in plasticity mechanisms can modify not only the synaptic weights, but also the dendritic synaptic composition to efficiently train large multi-compartment neurons. Transistor-level simulations demonstrate the functionality of the analog implementation and illustrate analogies to biological measurements.