Hyperparameter-Free Neurochaos Learning Algorithm for Classification

2025-08-02

AI Summary

Neurochaos Learning (NL) suffers from high computational complexity—requiring four-dimensional chaotic feature extraction per sample—and poor generalization due to sensitivity to multiple hyperparameters. To address these limitations, this paper proposes AutochaosNet: a training-free, hyperparameter-free neural chaos classification framework. Its core innovation lies in replacing conventional chaotic systems with a universal chaotic sequence derived from the Champernowne constant, coupled with input-driven spike-time boundary determination and a simplified Temporal-Multiplier (TM) / TM-Firing Rate (TM-FR) encoding scheme. This enables low-overhead, highly robust feature extraction. Experiments demonstrate that AutochaosNet achieves state-of-the-art (SOTA) accuracy across multi-class classification benchmarks, with near-zero training time, significantly reduced computational cost, and markedly improved generalization and scalability—establishing a novel paradigm for lightweight chaos-enabled intelligence.

Technology Category

• Machine Learning: Neuro-Symbolic Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

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 the 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.

Problem

• Eliminates hyperparameter tuning in Neurochaos Learning
• Reduces computational effort for feature extraction
• Improves classification performance without training

Innovation

• Hyperparameter-free Neurochaos Learning algorithm
• Universal chaotic sequence from Champernowne constant
• Input stimulus defines firing time bounds