Learning flow functions of spiking systems

Miguel Aguiar, Amritam Das, Karl H. Johansson
Proceedings of the 6th Annual Learning for Dynamics & Control Conference, PMLR 242:591-602, 2024.

Abstract

We propose a framework for surrogate modelling of spiking systems. These systems are often described by stiff differential equations with high-amplitude oscillations and multi-timescale dynamics, making surrogate models an attractive tool for system design.We parameterise the flow function of a spiking system in state-space using a recurrent neural network architecture, allowing for a direct continuous-time representation of the state trajectories which is particularly advantageous for this class of systems.The spiking nature of the signals makes for a data-heavy and computationally hard training process, and we describe two methods to mitigate these difficulties. We demonstrate our framework on two conductance-based models of biological neurons.

Cite this Paper

BibTeX
@InProceedings{pmlr-v242-aguiar24a, title = {Learning flow functions of spiking systems}, author = {Aguiar, Miguel and Das, Amritam and Johansson, Karl H.}, booktitle = {Proceedings of the 6th Annual Learning for Dynamics & Control Conference}, pages = {591--602}, year = {2024}, editor = {Abate, Alessandro and Cannon, Mark and Margellos, Kostas and Papachristodoulou, Antonis}, volume = {242}, series = {Proceedings of Machine Learning Research}, month = {15--17 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v242/aguiar24a/aguiar24a.pdf}, url = {https://proceedings.mlr.press/v242/aguiar24a.html}, abstract = {We propose a framework for surrogate modelling of spiking systems. These systems are often described by stiff differential equations with high-amplitude oscillations and multi-timescale dynamics, making surrogate models an attractive tool for system design.We parameterise the flow function of a spiking system in state-space using a recurrent neural network architecture, allowing for a direct continuous-time representation of the state trajectories which is particularly advantageous for this class of systems.The spiking nature of the signals makes for a data-heavy and computationally hard training process, and we describe two methods to mitigate these difficulties. We demonstrate our framework on two conductance-based models of biological neurons.} }
Endnote
%0 Conference Paper %T Learning flow functions of spiking systems %A Miguel Aguiar %A Amritam Das %A Karl H. Johansson %B Proceedings of the 6th Annual Learning for Dynamics & Control Conference %C Proceedings of Machine Learning Research %D 2024 %E Alessandro Abate %E Mark Cannon %E Kostas Margellos %E Antonis Papachristodoulou %F pmlr-v242-aguiar24a %I PMLR %P 591--602 %U https://proceedings.mlr.press/v242/aguiar24a.html %V 242 %X We propose a framework for surrogate modelling of spiking systems. These systems are often described by stiff differential equations with high-amplitude oscillations and multi-timescale dynamics, making surrogate models an attractive tool for system design.We parameterise the flow function of a spiking system in state-space using a recurrent neural network architecture, allowing for a direct continuous-time representation of the state trajectories which is particularly advantageous for this class of systems.The spiking nature of the signals makes for a data-heavy and computationally hard training process, and we describe two methods to mitigate these difficulties. We demonstrate our framework on two conductance-based models of biological neurons.
APA
Aguiar, M., Das, A. & Johansson, K.H.. (2024). Learning flow functions of spiking systems. Proceedings of the 6th Annual Learning for Dynamics & Control Conference, in Proceedings of Machine Learning Research 242:591-602 Available from https://proceedings.mlr.press/v242/aguiar24a.html.

Related Material