符号双注意力：在时间序列预测中捕捉符号依赖关系
Signed Dual Attention: Capturing Signed Dependencies in Time Series Forecasting
AI总结：
提出符号双注意力机制，通过双消息传递方案同时捕捉正负依赖关系，无需额外参数，提升时间序列预测性能。
摘要：
最初为自然语言处理开发的Transformer架构和注意力机制，现在已成为各种深度学习模型的核心，包括时间序列预测应用。然而，标准注意力机制隐含地假设同质性交互，限制了其对具有正负依赖关系（如时间序列）的数据建模能力。在这项工作中，我们引入了符号双注意力，一种新颖的注意力公式，无需额外参数即可捕捉正负关系模式。通过利用受相关结构启发的双消息传递方案，符号双注意力在单个共享块内传播支持和对比信息，有效实现了两个头注意力的表达能力而无需额外参数。该模块可以无缝集成到现有架构中，并在需要符号关系建模的某些情况下带来性能提升。这种方法为构建更具表达力和参数效率的Transformer开辟了道路。
英文摘要：
Initially developed for natural language processing, Transformer architectures and attention mechanisms are now central to a wide range of deep learning models, including applications in time series forecasting. A standard attention mechanism, however, implicitly assumes homophilic interactions, limiting its ability to model data with positive and negative dependencies, such as time series. In this work, we introduce the Signed Dual Attention, a novel attention formulation that captures both positive and negative relational patterns without additional parameters. By leveraging a dual message-passing scheme inspired by correlation structures, Signed Dual Attention propagates both supportive and contrastive information within a single shared block, effectively achieving the expressiveness of two head attention without additional parameters. This module can be seamlessly integrated into existing architectures and can yield performance gains in certain situations, requiring signed relational modeling. This approach opens a pathway toward more expressive and parameter-efficient transformers.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
