Neuromorphic chip simulates worm brain faster and more efficiently
Small-scale neural simulation of Caenorhabditis elegans on Loihi-2 using NeuroML2Loihi
Neural and Evolutionary Computing
Summary
Simulating brain activity in computers can be slow and use a lot of power. The authors created a tool called NeuroML2Loihi that helps run brain models on a special chip called Loihi 2, designed to mimic how real brains work. They tested this tool on a worm's nervous system model and showed it works much faster and uses less energy than traditional methods. This makes comparing brain simulations on different systems easier and could help neuroscience research.
What this means in practice
- •For neuromorphic engineers: Implement standardized biological neural models directly on Loihi 2 hardware using NeuroML2Loihi to enable low-latency, energy-efficient neuromorphic simulations.
- •For computational neuroscience platform developers: Integrate NeuroML-based models into neuromorphic software pipelines to compare and benchmark conventional and neuromorphic neural simulations efficiently.
Authors
Danny Rosen, Ramashish Gaurav, Yang Yi
Abstract
Neural simulation is an essential methodology for studying biological nervous systems, but it remains limited by the computational costs of repeatedly updating neural and synaptic states over many timesteps. Neuromorphic hardware is a promising substrate for this class of workloads because such devices combine event-driven execution, localized memory, and spike-based communication. However, neuromorphic \textit{software} remains comparatively fragmented and difficult to integrate with the model-description standards used by the computational neuroscience community. This disconnect limits reproducible experimentation and makes direct comparison between conventional and neuromorphic simulation workflows difficult. This paper introduces ``NeuroML2Loihi'', an open-source framework for translating NeuroML model descriptions into Loihi~2-deployable Lava primitives, enabling standardized biological models to be executed on Intel's Loihi~2 neuromorphic chip. We evaluate our framework using c302-derived \textit{C. elegans} neural models across three connectome-scale configurations. Cross-platform similarity is assessed using spike-train and voltage-trace metrics, while latency, power, and energy consumption are benchmarked against the conventional reference workflow. Across the evaluated models, Loihi~2 execution enabled through NeuroML2Loihi achieves 514--1211$\times$ lower latency and a 1.8--1.9$\times$ reduction in mean wattage relative to the NEURON reference, producing substantially lower total energy consumption primarily through reduced execution time. These results demonstrate that standards-based neuromorphic neural simulation is feasible on Loihi~2 and can support meaningful comparison against established computational neuroscience workflows.