Self-Training the Neurochaos Learning Algorithm

2026-01-03

arXiv.org

AI Summary

This work proposes a semi-supervised learning framework that integrates Neurochaos Learning with threshold-based self-training to address the limitations of conventional supervised learning in scenarios with scarce or imbalanced labeled data. By combining chaotic dynamics–driven feature extraction with a high-confidence pseudo-labeling mechanism, the approach effectively leverages limited labeled examples to enhance model performance. Evaluated on ten benchmark datasets using only 15% labeled data, the method significantly outperforms standard self-training baselines, achieving classification accuracy improvements of 188.66%, 158.58%, and 110.48% on the Iris, Wine, and Glass datasets, respectively. These results demonstrate a marked improvement in generalization and robustness under small-sample, nonlinear, and class-imbalanced conditions.

Technology Category

• Machine Learning: Semi-Supervised Learning
• Search and Optimization: Learning to Search
• 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
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

In numerous practical applications, acquiring substantial quantities of labelled data is challenging and expensive, but unlabelled data is readily accessible. Conventional supervised learning methods frequently underperform in scenarios characterised by little labelled data or imbalanced datasets. This study introduces a hybrid semi-supervised learning (SSL) architecture that integrates Neurochaos Learning (NL) with a threshold-based Self-Training (ST) method to overcome this constraint. The NL architecture converts input characteristics into chaos-based ring-rate representations that encapsulate nonlinear relationships within the data, whereas ST progressively enlarges the labelled set utilising high-confidence pseudo-labelled samples. The model's performance is assessed using ten benchmark datasets and five machine learning classifiers, with 85% of the training data considered unlabelled and just 15% utilised as labelled data. The proposed Self-Training Neurochaos Learning (NL+ST) architecture consistently attains superior performance gain relative to standalone ST models, especially on limited, nonlinear and imbalanced datasets like Iris (188.66%), Wine (158.58%) and Glass Identification (110.48%). The results indicate that using chaos-based feature extraction with SSL improves generalisation, resilience, and classification accuracy in low-data contexts.

Problem

• semi-supervised learning
• limited labelled data
• imbalanced datasets
• data scarcity
• supervised learning

Innovation

• Neurochaos Learning
• Self-Training
• Semi-Supervised Learning
• Chaos-based Feature Representation
• Pseudo-labeling