XAI-SOH-FL: Enhancing SOH-FL with Adaptive Aggregation and Explainable AI for Intrusion Detection in Heterogeneous IoT

2026-05-28 · Source: Artificial Intelligence · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cybersecurity & Data Privacy, Internet of Things (IoT) & Connected Devices · Depth: Expert, quick

Summary

The XAI-SOH-FL framework enhances the SOH-FL federated learning paradigm for intrusion detection in heterogeneous IoT environments by integrating adaptive aggregation and explainable AI. Addressing limitations like manual aggregation parameter tuning and lack of interpretability, XAI-SOH-FL introduces a dynamic γ selection mechanism based on similarity thresholding and employs Bayesian Optimization for automatic γ determination. It also incorporates SHAP (SHapley Additive exPlanations) to provide feature-level interpretability for intrusion detection decisions. Evaluated on the CICIDS2017 dataset, the proposed approach achieved an accuracy of 94.12% and an F1-score of 0.92, outperforming the baseline SOH-FL model with faster convergence. SHAP analysis highlighted Flow Duration and Packet Length as significant flow-level features influencing predictions, demonstrating an effective balance of accuracy, adaptability, and interpretability.

Key takeaway

For Machine Learning Engineers developing intrusion detection systems in heterogeneous IoT, XAI-SOH-FL offers a robust solution. You should consider integrating adaptive aggregation mechanisms and explainable AI techniques like SHAP to enhance model performance and interpretability. This approach can improve accuracy and F1-scores while reducing manual tuning efforts and providing crucial insights into detection decisions, especially with evolving data distributions.

Key insights

XAI-SOH-FL improves federated intrusion detection in IoT via adaptive aggregation and SHAP-based explainability.

Principles

• Dynamic aggregation adapts to data shifts.
• Bayesian Optimization automates parameter tuning.
• SHAP provides feature-level interpretability.

Method

XAI-SOH-FL integrates dynamic γ selection via similarity thresholding and Bayesian Optimization for aggregation, then uses SHAP for feature-level interpretability in federated intrusion detection.

In practice

• Apply dynamic γ for evolving data.
• Use SHAP to understand IDS decisions.
• Automate FL parameter tuning.

Topics

• Federated Learning
• Intrusion Detection
• Explainable AI
• IoT Security
• SHAP
• Adaptive Aggregation

Best for: Research Scientist, AI Scientist, Machine Learning Engineer, AI Security Engineer

Related on AIssential

• this 2019 ronny chieng bit aged into a prophecy — A 2019 comedy bit predicting AI replacing human thought has become a product category, highlighting rapid technological shifts.
• The Sequence Knowledge #870: Liquid Models and the Search for a Post-Transformer Architecture — Transformers excel at cloud-scale AI but are costly for on-device applications, where liquid models offer a dynamic, efficient alternative.
• Big Tech’s Looming Capability Crisis — AI's rapid advancement can swiftly render highly skilled human capabilities obsolete.
• The art and science of hyperparameter optimization on Amazon Nova Forge — Customizing LLMs on Amazon Nova Forge requires balancing strategic choices and systematic hyperparameter tuning to prevent catastrophic forgetting.
• Expressivity of congruence-based architectures for DNNs on positive-definite matrices — Semi-orthogonality in congruence layers limits DNN expressivity for SPD matrix classification due to spectral diversity loss.
• Error Bounds for a Diffusion Model-Based Drift Estimator — Theoretical guarantees for a diffusion model-based drift estimator in SDEs are now established through an explicit risk bound.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial Intelligence.