Reproduction of AdEx dynamics on neuromorphic hardware through data embedding and simulation-based inference

TL;DR

This paper demonstrates a method combining autoencoders and simulation-based inference to calibrate complex neuron models on neuromorphic hardware, enabling automated parameter estimation from membrane traces.

Contribution

It introduces a novel approach integrating autoencoders with SNPE for efficient model calibration on neuromorphic systems.

Findings

Autoencoder effectively extracts key features from membrane traces.

SNPE successfully approximates the posterior distribution of neuron parameters.

The combined method shows promise for optimizing complex physical system models.

Abstract

The development of mechanistic models of physical systems is essential for understanding their behavior and formulating predictions that can be validated experimentally. Calibration of these models, especially for complex systems, requires automated optimization methods due to the impracticality of manual parameter tuning. In this study, we use an autoencoder to automatically extract relevant features from the membrane trace of a complex neuron model emulated on the BrainScaleS-2 neuromorphic system, and subsequently leverage sequential neural posterior estimation (SNPE), a simulation-based inference algorithm, to approximate the posterior distribution of neuron parameters. Our results demonstrate that the autoencoder is able to extract essential features from the observed membrane traces, with which the SNPE algorithm is able to find an approximation of the posterior distribution. This suggests that the combination of an autoencoder with the SNPE algorithm is a promising optimization method for complex systems.