Mapping and Validating a Point Neuron Model on Intel's Neuromorphic Hardware Loihi
This work addresses the need for rigorous validation of neuromorphic hardware like Loihi to enable computational acceleration for neuroscience and AI research, though it is incremental as it focuses on validating an existing model on new hardware.
The researchers tackled the challenge of validating Intel's neuromorphic hardware Loihi for simulating brain models by implementing and testing Leaky Integrate and Fire models based on mouse primary visual cortex data. They found that Loihi replicates classical simulations efficiently and scales well in time and energy performance as networks grow larger, with notable improvements in scaling.
Neuromorphic hardware is based on emulating the natural biological structure of the brain. Since its computational model is similar to standard neural models, it could serve as a computational acceleration for research projects in the field of neuroscience and artificial intelligence, including biomedical applications. However, in order to exploit this new generation of computer chips, rigorous simulation and consequent validation of brain-based experimental data is imperative. In this work, we investigate the potential of Intel's fifth generation neuromorphic chip - `Loihi', which is based on the novel idea of Spiking Neural Networks (SNNs) emulating the neurons in the brain. The work is implemented in context of simulating the Leaky Integrate and Fire (LIF) models based on the mouse primary visual cortex matched to a rich data set of anatomical, physiological and behavioral constraints. Simulations on the classical hardware serve as the validation platform for the neuromorphic implementation. We find that Loihi replicates classical simulations very efficiently and scales notably well in terms of both time and energy performance as the networks get larger.