Complexity and Criticality in Neuro-Inspired Reservoirs

van der Vlag, Michiel; Morrison, Abigail; Jimenez-Romero, Cristian; Yegenoglu, Alper; Diaz, Sandra

doi:10.1007/978-3-032-04558-4_52

Journal Article/Contribution to a conference proceedings/Contribution to a book

FZJ-2025-04036

Complexity and Criticality in Neuro-Inspired Reservoirs

van der Vlag, M. (Corresponding author)FZJ* ; Yegenoglu, A. ; Jimenez-Romero, C.FZJ* ; Morrison, A.FZJ* ; Diaz, S.FZJ*

2026
Springer Nature Switzerland Cham
ISBN: 978-3-032-04557-7, 978-3-032-04558-4 (electronic)

[Ebook] Artificial Neural Networks and Machine Learning – ICANN 2025 : 34th International Conference on Artificial Neural Networks, Kaunas, Lithuania, September 9–12, 2025, Proceedings, Part I / Senn, Walter ; Sanguineti, Marcello ; Saudargiene, Ausra ; Tetko, Igor ; Villa, Alessandro E. P. ; Jirsa, Viktor K. ; Bengio, Yoshua 1st ed. 2026, Cham : Springer Nature Switzerland, 2026,
34th International Conference on Artificial Neural Networks, ICANN, Kaunas, Lithuania, 9 Sep 2025 - 12 Sep 2025 Artificial Neural Networks and Machine Learning – ICANN 2025 / Senn, Walter (Editor) [https://orcid.org/0000-0003-3622-0497] ; Cham : Springer Nature Switzerland, 2026, Chapter 52 ; ISSN: 0302-9743=1611-3349 ; ISBN: 978-3-032-04557-7=978-3-032-04558-4 ; doi:10.1007/978-3-032-04558-4 16068(LNCS), : 1st ed. 2026, 645 - 657 (2026) [10.1007/978-3-032-04558-4_52]

This record in other databases:

Please use a persistent id in citations: doi:10.1007/978-3-032-04558-4_52 doi:10.34734/FZJ-2025-04036

Abstract: Understanding information propagation and computational capabilities in large-scale brain models is crucial for advancing both neuroscience and neuro-inspired computing. It remains unknown how complexity affects the performance of simulated biological networks in predicting non-linear dynamics. Here, this issue is addressed by integrating reservoir computing, a recurrent neural network architecture, with The Virtual Brain, a whole-brain simulation platform, to create amore neurophysiologically-plausible machine learning framework. Metrics derived from nonlinear dynamics and complexity theory, including the largest Lyapunov exponent, which captures chaotic behavior, Detrended Fluctuation Analysis, which assesses temporal correlations, and the Perturbational Complexity Index, which evaluates the complexity of neural responses, provide a quantitative framework for characterizing cognitive dynamics. Deploying this framework on High Performance Computing enables a thorough exploration of the vast parameter space, utilizing a diverse evaluation framework that assesses simulations through these metrics, implemented on Graphics Processing Units (GPUs). This enables the identification of optimal parameter regimes, a comprehensive characterization of the complex dynamics exhibited by the system, and a deeper understanding of the underlying mechanisms governing the TVB-based reservoir’s computational capabilities. Ridge regression, accelerated also by GPUs, is used to extract the predictive capacity from the reservoir states. The results suggest that edge-of-chaosdynamics correspond to enhanced memory and prediction accuracy, supporting the potential of TVB-based reservoirs for brain-inspired machine learning.

Keyword(s): Artificial intelligence (LCSH) ; Computers (LCSH) ; Application software (LCSH) ; Computer networks (LCSH)

Classification:

ddc:006.3

Note: This preprint has not undergone peer review (when applicable) or any post-submission improvements or corrections. The Version of Record of this contribution is published in "Artificial Neural Networks and Machine Learning – ICANN 2025", and is available online at https://link.springer.com/book/10.1007/978-3-032-04558-4

Contributing Institute(s):