[论文解读] Gated Adaptation for Continual Learning in Human Activity Recognition

Reza Rahimi Azghan, Gautham Krishna Gudur+5 more|arXiv (Cornell University)|Mar 8, 2026

Context-Aware Activity Recognition Systems被引用 0

一句话总结

本文提出一种用于人类活动识别（HAR）的参数高效的连续学习方法，该方法在冻结的预训练骨干网基础上使用通道级门控调制，降低遗忘并提高准确度，训练参数占比不到2%。

ABSTRACT

Wearable sensors in Internet of Things (IoT) ecosystems increasingly support applications such as remote health monitoring, elderly care, and smart home automation, all of which rely on robust human activity recognition (HAR). Continual learning systems must balance plasticity (learning new tasks) with stability (retaining prior knowledge), yet AI models often exhibit catastrophic forgetting, where learning new tasks degrades performance on earlier ones. This challenge is especially acute in domain-incremental HAR, where on-device models must adapt to new subjects with distinct movement patterns while maintaining accuracy on prior subjects without transmitting sensitive data to the cloud. We propose a parameter-efficient continual learning framework based on channel-wise gated modulation of frozen pretrained representations. Our key insight is that adaptation should operate through feature selection rather than feature generation: by restricting learned transformations to diagonal scaling of existing features, we preserve the geometry of pretrained representations while enabling subject-specific modulation. We provide a theoretical analysis showing that gating implements a bounded diagonal operator that limits representational drift compared to unconstrained linear transformations. Empirically, freezing the backbone substantially reduces forgetting, and lightweight gates restore lost adaptation capacity, achieving stability and plasticity simultaneously. On PAMAP2 with 8 sequential subjects, our approach reduces forgetting from 39.7% to 16.2% and improves final accuracy from 56.7% to 77.7%, while training less than 2% of parameters. Our method matches or exceeds standard continual learning baselines without replay buffers or task-specific regularization, confirming that structured diagonal operators are effective and efficient under distribution shift.

研究动机与目标

• 在边缘设备上解决领域增量HAR中的灾难性遗忘问题。
• 在最小化可训练参数的前提下实现稳定性-塑性权衡。
• 在门控自适应下提供对表示漂移有界的理论保证。
• 在存在被试分布漂移的多项HAR基准数据集上证明有效性。
• 强调在设备上进行连续学习时无需数据回放的实用性。

提出的方法

• 冻结预训练的HAR骨干网，在每个骨干块后增加轻量级的通道级门控。
• 通过全局池化得到通道描述符，经过瓶颈激励生成门控 g_l，并对 U_l 施加对角缩放 D(g_l)。
• 仅训练门控参数和一个共享分类器，而骨干网保持固定。
• 使用跨任务的共享线性分类器，并通过时序池化得到最终的对数概率（logits）。
• 给出理论分析，表明门控会诱导一个有界对角运算符，从而限制漂移。

Figure 1: Catastrophic forgetting in subject-incremental HAR: accuracy on Subject 1 drops from 85% to 40% after training on just three additional subjects (PAMAP2 dataset).

实验结果

研究问题

• RQ1 在冻结骨干网的前提下，通道级对角门控能否为受被试影响的 HAR 的稳定连续自适应提供支持？
• RQ2 在遗忘和最终准确度方面，门控自适应相较全量微调或最小适应有何差异？在HAR基准上？
• RQ3 在领域增量学习下，门控自适应对特征漂移和对数概率稳定性的理论保障是什么？
• RQ4 达到在新HAR被试上具有竞争力的性能，需要训练多少比例的模型参数？
• RQ5 在HAR数据集中观察到的通道级领域漂移下，所提出的门控是否足够？

主要发现

• 对于 PAMAP2 的8个连续被试，遗忘从 39.7%（可训练骨干网）降至 16.2%。
• 最终准确度从 56.7% 提升至 77.7%。
• 训练参数占比低于 2%。
• 基于门控的自适应在没有回放缓冲或任务特定正则化的情况下，与或超越标准连续学习基线。
• 在三个 HAR 基准数据集（PAMAP2、UCI-HAR、DSA）的领域增量设置下，实验显示出对被试数量增加（最多30个）的持续改进。

Figure 2: Overview of the proposed continual learning system for HAR, comprising data collection, preprocessing, and a classification model with a frozen pretrained backbone modulated by lightweight trainable gates.