Lite-RVFL: A Lightweight Random Vector Functional-Link Neural Network for Learning Under Concept Drift

2025-06-09

AI Summary

To address the high computational overhead and poor real-time adaptability of existing online learning methods under concept drift—particularly those requiring frequent retraining or explicit drift detection—this paper proposes Lite-RVFL, a lightweight adaptive model built upon the Random Vector Functional Link (RVFL) architecture. Lite-RVFL introduces, for the first time, an exponentially increasing time-weighted objective function, with theoretical proof of asymptotic adaptivity under concept drift. It further designs a novel incremental update mechanism that eliminates both retraining and explicit drift detection, drastically reducing computational complexity. Evaluated on real-world security assessment tasks, Lite-RVFL achieves faster drift response and higher prediction accuracy compared to state-of-the-art approaches, while reducing computational cost by an order of magnitude. The method effectively captures dynamic temporal patterns in streaming data, demonstrating strong practical viability for resource-constrained, real-time applications.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Models for Web evolution

Abstract

The change in data distribution over time, also known as concept drift, poses a significant challenge to the reliability of online learning methods. Existing methods typically require model retraining or drift detection, both of which demand high computational costs and are often unsuitable for real-time applications. To address these limitations, a lightweight, fast and efficient random vector functional-link network termed Lite-RVFL is proposed, capable of adapting to concept drift without drift detection and retraining. Lite-RVFL introduces a novel objective function that assigns weights exponentially increasing to new samples, thereby emphasizing recent data and enabling timely adaptation. Theoretical analysis confirms the feasibility of this objective function for drift adaptation, and an efficient incremental update rule is derived. Experimental results on a real-world safety assessment task validate the efficiency, effectiveness in adapting to drift, and potential to capture temporal patterns of Lite-RVFL. The source code is available at https://github.com/songqiaohu/Lite-RVFL.

Problem

• Addresses concept drift in online learning efficiently
• Eliminates need for drift detection and retraining
• Adapts to data distribution changes in real-time

Innovation

• Lightweight RVFL network for concept drift
• Exponential weighting for recent data emphasis
• Incremental update without retraining or detection