White-Box AI Model: Next Frontier of Wireless Communications

2025-04-12

AI Summary

To address critical limitations of black-box AI models in wireless communications—including poor interpretability, lack of mathematical verifiability, and limited physical-layer transferability—this paper proposes the White-box AI (WAI) paradigm. WAI integrates theory-driven causal modeling with verifiable optimization-path design to reinterpret DNN/Transformer architectures, enabling co-adaptation among optimization objectives, network structure, and physical-layer tasks. It establishes, for the first time, a systematic methodology for white-box AI tailored to wireless communications, overcoming the fundamental bottlenecks of theoretical uninterpretability and non-traceable optimization inherent in conventional black-box approaches for signal processing and resource allocation. Experimental results demonstrate that WAI significantly outperforms black-box models in interpretability, mathematical verifiability, and cross-scenario physical-layer transferability. This work provides a foundational methodological framework for building trustworthy intelligent wireless systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Machine Learning: Hardware-aware ML
• Philosophy and Ethics of AI: Safety, Robustness & Trustworthiness

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Security and privacy of machine learning and AI applications
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

White-box AI (WAI), or explainable AI (XAI) model, a novel tool to achieve the reasoning behind decisions and predictions made by the AI algorithms, makes it more understandable and transparent. It offers a new approach to address key challenges of interpretability and mathematical validation in traditional black-box models. In this paper, WAI-aided wireless communication systems are proposed and investigated thoroughly to utilize the promising capabilities. First, we introduce the fundamental principles of WAI. Then, a detailed comparison between WAI and traditional black-box model is conducted in terms of optimization objectives and architecture design, with a focus on deep neural networks (DNNs) and transformer networks. Furthermore, in contrast to the traditional black-box methods, WAI leverages theory-driven causal modeling and verifiable optimization paths, thereby demonstrating potential advantages in areas such as signal processing and resource allocation. Finally, we outline future research directions for the integration of WAI in wireless communication systems.

Problem

• Develops white-box AI for transparent wireless communication decisions
• Compares WAI and black-box models in optimization and design
• Explores WAI's advantages in signal processing and resource allocation

Innovation

• White-box AI enhances interpretability in wireless systems
• Theory-driven causal modeling replaces black-box methods
• Verifiable optimization paths improve signal processing