用异构超图学习提升真实场景下人体活动识别准确率
Deep Heterogeneous Contrastive Hyper-Graph Learning for In-the-Wild Context-Aware Human Activity Recognition
- 构建三类子超图，分层处理不同传感器数据的异构性
- 对比损失函数提升节点异构性建模能力，性能提升5.8%~16.7%
- 适用于多设备、多场景下的复杂活动识别任务
人体活动识别（HAR）是具有挑战性的多标签分类问题，因为活动可能同时发生，且同一活动在不同上下文（如设备放置位置不同）下的传感器信号存在差异。本文提出一种深度异构对比超图学习框架（DHC-HGL），以消息传递和邻域聚合的方式捕捉真实场景下上下文感知活动识别（CA-HAR）的异构超图特性。现有工作仅研究同质或浅层节点异构图。DHC-HGL通过创新性地：1）构建三类不同类型的子超图，并分别通过针对边异构性设计的自定义超图卷积（HGC）层处理；2）采用对比损失函数确保节点异构性建模。在两个CA-HAR数据集上的严格评估表明，DHC-HGL在马修斯相关系数（MCC）上相比最优基线提升5.8%至16.7%，宏平均F1得分提升3.0%至8.4%。同时通过UMAP可视化学习到的节点嵌入，增强模型可解释性。
原文摘要 · Abstract (English)
Human Activity Recognition (HAR) is a challenging, multi-label classification problem as activities may co-occur and sensor signals corresponding to the same activity may vary in different contexts (e.g., different device placements). This paper proposes a Deep Heterogeneous Contrastive Hyper-Graph Learning (DHC-HGL) framework that captures heterogenous Context-Aware HAR (CA-HAR) hypergraph properties in a message-passing and neighborhood-aggregation fashion. Prior work only explored homogeneous or shallow-node-heterogeneous graphs. DHC-HGL handles heterogeneous CA-HAR data by innovatively 1) Constructing three different types of sub-hypergraphs that are each passed through different custom HyperGraph Convolution (HGC) layers designed to handle edge-heterogeneity and 2) Adopting a contrastive loss function to ensure node-heterogeneity. In rigorous evaluation on two CA-HAR datasets, DHC-HGL significantly outperformed state-of-the-art baselines by 5.8% to 16.7% on Matthews Correlation Coefficient (MCC) and 3.0% to 8.4% on Macro F1 scores. UMAP visualizations of learned CA-HAR node embeddings are also presented to enhance model explainability.
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。