将波干涉原理融入自注意力，提升复数信号建模能力
Holographic Transformers for Complex-Valued Signal Processing: Integrating Phase Interference into Self-Attention
- 用相位差调制交互，实现相干叠加的复数注意力机制
- 在极化SAR分类与无线信道预测中达高精度与低误差
- 适合处理雷达、通信等需精确相位信息的任务
复数信号同时包含幅度和相位信息，但多数深度模型将注意力视为实值相关，忽略了干涉效应。我们提出全息变压器（Holographic Transformer），一种受物理启发的架构，将波干涉原理融入自注意力。全息注意力通过相对相位调制交互，并对值进行相干叠加，确保幅度与相位的一致性。双头解码器同时重建输入并预测任务输出，防止损失函数侧重幅度时出现相位坍缩。实验表明，全息注意力实现了离散干涉算子，且在线性混叠下保持相位一致性。在极化合成孔径雷达（PolSAR）图像分类和无线信道预测任务中表现优异，分类准确率高，F1得分高，回归误差低，对相位扰动更具鲁棒性。结果表明，强制注意力的物理一致性可带来复数学习的通用提升，并提供统一的相干信号建模范式。代码已开源：https://github.com/EonHao/Holographic-Transformers。
原文摘要 · Abstract (English)
Complex-valued signals encode both amplitude and phase, yet most deep models treat attention as real-valued correlation, overlooking interference effects. We introduce the Holographic Transformer, a physics-inspired architecture that incorporates wave interference principles into self-attention. Holographic attention modulates interactions by relative phase and coherently superimposes values, ensuring consistency between amplitude and phase. A dual-headed decoder simultaneously reconstructs the input and predicts task outputs, preventing phase collapse when losses prioritize magnitude over phase. We demonstrate that holographic attention implements a discrete interference operator and maintains phase consistency under linear mixing. Experiments on PolSAR image classification and wireless channel prediction show strong performance, achieving high classification accuracy and F1 scores, low regression error, and increased robustness to phase perturbations. These results highlight that enforcing physical consistency in attention leads to generalizable improvements in complex-valued learning and provides a unified, physics-based framework for coherent signal modeling. The code is available at https://github.com/EonHao/Holographic-Transformers.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。