Hyperparameter-Free Neurochaos Learning Algorithm for Classification

Henry, Akhila and Nagaraj, Nithin (2025) Hyperparameter-Free Neurochaos Learning Algorithm for Classification. arXiv:2508.01478 [cs.LG].

Text
Hyperparameter-Free Neurochaos Learning Algorithm for Classification.pdf - Published Version
Download (2MB)

Contribution	Name	Email

Abstract:	Neurochaos Learning (NL) is a brain-inspired classification framework that employs chaotic dynamics to extract features from input data and yields state-of-the-art performance on classification tasks. However, NL requires the tuning of multiple hyperparameters and computing of four chaotic features per input sample. In this paper, we propose AutochaosNet – a novel, hyperparameter-free variant of the NL algorithm that eliminates the need for both training and parameter optimization. AutochaosNet leverages a universal chaotic sequence derived from the Champernowne constant and uses the input stimulus to define firing time bounds for feature extraction. Two simplified variants—TM AutochaosNet and TM-FR AutochaosNet—are evaluated against existing NL architecture - ChaosNet. Our results demonstrate that AutochaosNet achieves competitive or superior classification performance while significantly reducing training time due to reduced computational effort. In addition to eliminating training and hyperparameter tuning, AutochaosNet exhibits excellent generalisation capabilities, making it a scalable and efficient choice for real-world classification tasks. Future work will focus on identifying universal orbits under various chaotic maps and incorporating them into the NL framework to further enhance performance.
Item Type:	Journal Paper
Subjects:	School of Natural and Engineering Sciences > Nonlinear Dynamics
Divisions:	Schools > Natural Sciences and Engineering
Date Deposited:	27 Apr 2026 13:42
Last Modified:	27 Apr 2026 13:42
Official URL:	https://arxiv.org/abs/2508.01478
Related URLs:
Funders:	*
Projects:	*
DOI:	https://doi.org/10.48550/arXiv.2508.01478
URI:	http://eprints.nias.res.in/id/eprint/3322

Actions (login required)

View Item