[论文解读] Transfer Learning in Human Activity Recognition: A Survey

Sourish Gunesh Dhekane, Thomas Ploetz|arXiv (Cornell University)|Jan 18, 2024

Context-Aware Activity Recognition Systems被引用 9

一句话总结

本综述分析基于传感器的 HAR 的迁移学习方法，聚焦智能家居和可穿戴设备，并将挑战映射到解决方案以指导未来研究。

ABSTRACT

Sensor-based human activity recognition (HAR) has been an active research area, owing to its applications in smart environments, assisted living, fitness, healthcare, etc. Recently, deep learning based end-to-end training has resulted in state-of-the-art performance in domains such as computer vision and natural language, where large amounts of annotated data are available. However, large quantities of annotated data are not available for sensor-based HAR. Moreover, the real-world settings on which the HAR is performed differ in terms of sensor modalities, classification tasks, and target users. To address this problem, transfer learning has been employed extensively. In this survey, we focus on these transfer learning methods in the application domains of smart home and wearables-based HAR. In particular, we provide a problem-solution perspective by categorizing and presenting the works in terms of their contributions and the challenges they address. We also present an updated view of the state-of-the-art for both application domains. Based on our analysis of 205 papers, we highlight the gaps in the literature and provide a roadmap for addressing them. This survey provides a reference to the HAR community, by summarizing the existing works and providing a promising research agenda.

研究动机与目标

• 定义在传感器模态和领域/任务变体下的 HAR 及 HAR 情境中的迁移学习。
• 将用于 HAR 的迁移学习方法按问题与解决方案视角进行分类。
• 评述智能家居和可穿戴 HAR 的前沿迁移学习方法。
• 找出空白并提出未来 HAR 迁移学习研究的路线图。

提出的方法

• 使用领域与任务形式化来呈现 HAR 中迁移学习的问题与解决方案视角。
• 将方法分为实例、特征、参数和知识库迁移。
• 讨论异质迁移、任务差异及相应的映射/变换。
• 总结智能家居和可穿戴设备的数据集、挑战及应用特定考量。
• 提供路线图并识别尚未解决的挑战和潜在方法。

实验结果

研究问题

• RQ1 在 HAR 的迁移学习中，核心问题设定是什么（领域与任务关系）？
• RQ2 哪些解空间策略（实例、特征、参数、知识库迁移）被用于实现 HAR 迁移学习？
• RQ3 迁移学习方法如何应对智能家居和可穿戴 HAR 的异质性和任务差异？
• RQ4 当前 HAR 迁移学习研究存在哪些空白，提出了哪些解决空白的路线图？

主要发现

• HAR 迁移学习面临特征空间异质性、标注数据有限、噪声和传感器异质性等挑战。
• 解决方案包括源/目标特征之间的映射、到公共空间的变换，以及利用元特征/语义进行迁移。
• 智能家居 HAR 常涉及具有不同传感器布局和标签映射的异构迁移，需要专门的对齐策略。
• 可穿戴 HAR 也面临设备异质性和分布漂移，催生如神经嵌入和自编码器等迁移方法。
• 本综述覆盖超过205篇工作，提供问题-解决方案视角及解决尚未解决挑战的路线图。