Simulation-based Inference for model parameterization on analog neuromorphic hardware

• Verfasst von: Kaiser, Jakob [VerfasserIn]
• Verfasst von: Stock, Raphael [VerfasserIn]
• Verfasst von: Müller, Eric [VerfasserIn]
• Verfasst von: Schemmel, Johannes [VerfasserIn]
• Verfasst von: Schmitt, Sebastian [VerfasserIn]
• Titel: Simulation-based Inference for model parameterization on analog neuromorphic hardware
• Verf.angabe: Jakob Kaiser, Raphael Stock, Eric Müller, Johannes Schemmel, and Sebastian Schmitt
• E-Jahr: 2023
• Jahr: August 27, 2023
• Umfang: 10 S.
• Fussnoten: Gesehen am 02.10.2023
• Titel Quelle: Enthalten in: De.arxiv.org
• Ort Quelle: [Erscheinungsort nicht ermittelbar] : Arxiv.org, 1991
• Jahr Quelle: 2023
• Band/Heft Quelle: (2023), Artikel-ID 2303.16056, Seite 1-10
• DOI: doi:10.48550/arXiv.2303.16056
• Datenträger: Online-Ressource
• Sprache: eng
• Bibliogr. Hinweis: Forschungsdaten: Kaiser, Jakob: Simulation-based inference for model parameterization on analog neuromorphic hardware [data]
• Sach-SW: Computer Science - Neural and Evolutionary Computing
• K10plus-PPN: 1860620604

Abstract

The BrainScaleS-2 (BSS-2) system implements physical models of neurons as well as synapses and aims for an energy-efficient and fast emulation of biological neurons. When replicating neuroscientific experiment results, a major challenge is finding suitable model parameters. This study investigates the suitability of the sequential neural posterior estimation (SNPE) algorithm for parameterizing a multi-compartmental neuron model emulated on the BSS-2 analog neuromorphic hardware system. In contrast to other optimization methods such as genetic algorithms or stochastic searches, the SNPE algorithms belongs to the class of approximate Bayesian computing (ABC) methods and estimates the posterior distribution of the model parameters; access to the posterior allows classifying the confidence in parameter estimations and unveiling correlation between model parameters. In previous applications, the SNPE algorithm showed a higher computational efficiency than traditional ABC methods. For our multi-compartmental model, we show that the approximated posterior is in agreement with experimental observations and that the identified correlation between parameters is in agreement with theoretical expectations. Furthermore, we show that the algorithm can deal with high-dimensional observations and parameter spaces. These results suggest that the SNPE algorithm is a promising approach for automating the parameterization of complex models, especially when dealing with characteristic properties of analog neuromorphic substrates, such as trial-to-trial variations or limited parameter ranges.