Enhancing IoT Network Security through Adaptive Curriculum Learning and XAI

TL;DR

This study presents a scalable and lightweight Curriculum Learning framework enhanced with Explainable AI (XAI) techniques, like LIME, to ensure transparency and adaptability and establishes this framework as a robust, transparent, and high-performance solution for IoT network security.

Abstract

To address the critical need for secure IoT networks, this study presents a scalable and lightweight curriculum learning framework enhanced with Explainable AI (XAI) techniques, including LIME, to ensure transparency and adaptability. The proposed model employs novel neural network architecture utilized at every stage of Curriculum Learning to efficiently capture and focus on both short- and long-term temporal dependencies, improve learning stability, and enhance accuracy while remaining lightweight and robust against noise in sequential IoT data. Robustness is achieved through staged learning, where the model iteratively refines itself by removing low-relevance features and optimizing performance. The workflow includes edge-optimized quantization and pruning to ensure portability that could easily be deployed in the edge-IoT devices. An ensemble model incorporating Random Forest, XGBoost, and the staged learning base further enhances generalization. Experimental results demonstrate 98% accuracy on CIC-IoV-2024 and CIC-APT-IIoT-2024 datasets and 97% on EDGE-IIoT, establishing this framework as a robust, transparent, and high-performance solution for IoT network security.

Figures (5)

• Figure 1: The workflow of the Proposed Framework
• Figure 2: Layer Vs Parameters In Neural Network
• Figure 3: Accuracy Improvements by Integrating Key Components in the Proposed Neural Network Structure
• Figure 4: Hyper-Parameters Importance in the Proposed Neural Network Structure
• Figure 5: Example LIME Explanation for Attack Classification, Highlighting Key Features and Their Contributions to the Model's Prediction