X-REFINE: XAI-based RElevance input-Filtering and archItecture fiNe-tuning for channel Estimation

2026-02-25

AI Summary

This work addresses the black-box nature and high computational complexity of deep learning-based approaches for 6G channel estimation by proposing the X-REFINE framework. X-REFINE introduces, for the first time, the symbol-stable LRP-ε backpropagation rule into end-to-end explainability-aware optimization, enabling simultaneous assessment of the importance of input subcarriers and hidden neurons. Leveraging high-resolution relevance scores, the method jointly performs input feature filtering and fine-grained network architecture pruning. Experimental results demonstrate that X-REFINE significantly reduces computational complexity across diverse scenarios while maintaining robust bit error rate performance, effectively achieving a balanced trade-off among interpretability, accuracy, and efficiency.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Search and Optimization: Learning to Search
• Computer Vision: Learning & Optimization for CV

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

AI-native architectures are vital for 6G wireless communications. The black-box nature and high complexity of deep learning models employed in critical applications, such as channel estimation, limit their practical deployment. While perturbation-based XAI solutions offer input filtering, they often neglect internal structural optimization. We propose X-REFINE, an XAI-based framework for joint input-filtering and architecture fine-tuning. By utilizing a decomposition-based, sign-stabilized LRP epsilon rule, X-REFINE backpropagates predictions to derive high-resolution relevance scores for both subcarriers and hidden neurons. This enables a holistic optimization that identifies the most faithful model components. Simulation results demonstrate that X-REFINE achieves a superior interpretability-performance-complexity trade-off, significantly reducing computational complexity while maintaining robust bit error rate (BER) performance across different scenarios.

Problem

• channel estimation
• XAI
• model interpretability
• computational complexity
• 6G wireless communications

Innovation

• XAI
• channel estimation
• architecture fine-tuning
• input filtering
• LRP