基于智能手机的人类活动识别的域泛化：组件与交互的系统分析
Domain Generalization for Smartphone-Based Human Activity Recognition: A Systematic Analysis of Components and Interactions
- Institute of Computing（计算研究所）
- University of Campinas（坎皮纳斯大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究系统分析智能手机HAR中域泛化各组件的独立与联合效果，发现联合配置具互补性，且模型选择策略对实现性能至关重要。
AI中文摘要：
基于智能手机的人类活动识别（HAR）模型在用户、设备、传感器放置位置、环境和采集协议变化引起的分布偏移下，性能往往会下降。域泛化（DG）通过从源域学习而不访问目标数据来解决这一问题。现有的DG方法涵盖训练目标、表示初始化和架构修改，但这些组件通常被孤立地评估，尽管它们在训练流程的不同阶段发挥作用。我们提出了一个大规模、受控的智能手机HAR域泛化基准，包含超过410,000次实验，涵盖四种模型架构、十三种训练目标（包括经验风险最小化（ERM））、五种初始化策略、四种架构配置以及两种偏移场景：跨数据集和跨位置。结果表明，单个DG组件提供的增益有限且高度依赖条件。替代目标很少能持续优于ERM，自监督初始化在特定设置中有帮助，而架构修改，特别是动态域泛化，提供了最明显的独立改进。然而，联合配置通常优于其单个组件，并表现出互补性，有时甚至是超加性的交互，尽管增益仍依赖于模型和偏移类型。类别层面的分析表明，最强的配置主要改善了困难、对偏移敏感的决策边界。最后，oracle检查点分析揭示了大量未实现的性能：源验证选择在跨数据集和跨位置设置中分别仅恢复了可用oracle增益的53%和26%。总体而言，有效的HAR域泛化需要联合设计DG组件和稳健的模型选择策略。
英文摘要：
Smartphone-based Human Activity Recognition (HAR) models often degrade under distribution shifts caused by changes in users, devices, sensor placements, environments, and acquisition protocols. Domain Generalization (DG) addresses this problem by learning from source domains without access to target data. Existing DG methods span training objectives, representation initialization, and architectural modifications, but these components are typically evaluated in isolation despite operating at different stages of the learning pipeline. We present a large-scale controlled benchmark of DG for smartphone-based HAR, comprising more than 410,000 experiments across four model architectures, thirteen training objectives including Empirical Risk Minimization (ERM), five initialization strategies, four architectural configurations, and two shift scenarios: cross-dataset and cross-position. Results show that individual DG components provide limited and highly conditional gains. Alternative objectives rarely outperform ERM consistently, self-supervised initialization helps in specific settings, and architectural modifications, particularly Dynamic Domain Generalization, provide the clearest standalone improvements. Joint configurations, however, frequently outperform their individual components and exhibit complementary and sometimes super-additive interactions, although gains remain model- and shift-dependent. Class-level analysis shows that the strongest configurations mainly improve difficult, shift-sensitive decision boundaries. Finally, oracle checkpoint analysis reveals substantial unrealized performance: source-validation selection recovers only 53% and 26% of the available oracle gain in cross-dataset and cross-position settings, respectively. Overall, effective HAR domain generalization requires jointly designing DG components and robust model-selection strategies.
