Evaluating topological fitness of human brain-inspired sub-circuits in Echo State Networks

Bach Viet Nguyen, Tianlong Chen, Shu Yang, Bojian Hou, Li Shen, Duy Duong-Tran

Keywords: Computational Neuroscience, Neural Data Analysis, Neuromorphic Computing, Recurrent Neural Networks, Echo-state Networks, Reservoir Computing, Network Topology, Bio-inspired Neural Networks

Abstract:

Recent years have witnessed an emerging trend in neuromorphic computing that centers around the use of brain connectomics as a blueprint for artificial neural networks. Connectomics-based neuromorphic computing has primarily focused on embedding human brain large-scale structural connectomes (SCs), as estimated from diffusion Magnetic Resonance Imaging (dMRI) modality, to echo-state networks (ESNs). A critical step in ESN embedding requires pre-determined read-in and read-out layers constructed by the induced subgraphs (e.g., a priori set of functional sub-circuits/networks) of the embedded reservoir (e.g., SCs). As a priori set of functional sub-circuits are derived from functional MRI (fMRI) modality, it is unknown, till this point, whether the embedding of fMRI-induced sub-circuits/networks onto SCs is well justified from i) the neuro-physiological perspective and ii) ESN performance across a variety of tasks. In this paper, we proposed a pipeline to implement and evaluate ESNs with various embedded topology and processing/memorization tasks. To this end, we showed that different performance optimums are highly dependent on the neuro-physiological characteristics of these pre-determined fMRI-induced sub-circuits. In general, fMRI-induced sub-circuit-embedded ESN outperforms simple bipartite and various null models with feed-forward properties commonly seen in MLP for different tasks and reservoir criticality conditions. Noticeably, we found that the reservoir model performance is heavily dependent on the functional sub-circuits neuro-physiological properties with respect to different cognitive tasks and their corresponding computation-memorization balances. Specifically, we showed that default mode network's superior performance across the majority of tasks is related to its functional dichotomy property. Finally, we provided a thorough analysis of the topological properties of pre-determined fMRI-induced sub-circuits and highlighted their graph-theoretical properties that play significant roles in determining the ESN performance.

Primary Area: neurosymbolic & hybrid AI systems (physics-informed, logic & formal reasoning, etc.)

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.

Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.

Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.

Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.

No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.

Submission Number: 7479