Abstract:Abstract Human activity recognition (HAR), which aims at inferring the behavioral patterns of people, is a fundamental research problem in digital health and ambient intelligence. The application of machine learning methods in HAR has been investigated vigorously in recent years. However, there are still a number of challenges confronting the task, where one significant barrier lies in the longstanding shortage of annotations. To address this issue, we establish a new paradigm for HAR, which integrates active learning and semi-supervised learning into one framework. The main idea is to reduce the annotation cost by actively selecting the most informative samples for annotation, as well as leveraging the unlabelled instances in a semi-supervised way. In particular, we propose to utilize the massive unlabelled data via temporal ensembling of convolutional neural networks (CNN), which yields robust consensus predictions by aggregating the outputs of the training networks on different epochs. We conducted extensive experiments on three public benchmark datasets. The proposed method achieves Macro F1 values of 0.76, 0.45 and 0.91 in a low annotation scenario on PAMAP2, USCHAD and UCIHAR datasets respectively, outperforming a multitude of state-of-the-art deep models. The ablation study proves the effectiveness of the two components of the framework, i.e., active learning-based sample selection and semi-supervised model training with temporal ensembling, in alleviating the issue of insufficient labels. Cross-validation and statistical significance experiments further demonstrate the robustness and generalization ability of the proposed method. The source codes are available at https://github.com/HaixiaBi1982/ActSemiCNNAct.

Weakly Supervised Human Activity Recognition from Wearable Sensors by Recurrent Attention Learning

Attention-based Convolutional Neural Network for Weakly Labeled Human Activities Recognition with Wearable Sensors

Lstm With Uniqueness Attention For Human Activity Recognition

A Semisupervised Recurrent Convolutional Attention Model for Human Activity Recognition

Attention-based LSTM Network for Wearable Human Activity Recognition

Consistency Based Weakly Self-Supervised Learning for Human Activity Recognition with Wearables

Context-aware mutual learning for semi-supervised human activity recognition using wearable sensors

Weakly Supervised Multi-Task Representation Learning for Human Activity Analysis Using Wearables

Deep Residual Bidir-LSTM for Human Activity Recognition Using Wearable Sensors

Research on Human Activity Recognition Based on Active Learning

Activation‐based recurrent learning method for wearable sensor data processing in human activity recognition

Deep Convolutional and LSTM Recurrent Neural Networks for Multimodal Wearable Activity Recognition

Multi-STMT: Multi-Level Network for Human Activity Recognition Based on Wearable Sensors

Deep, Convolutional, and Recurrent Models for Human Activity Recognition using Wearables

An active semi-supervised deep learning model for human activity recognition

Human activity recognition using wearable sensors by heterogeneous convolutional neural networks

Unsupervised Statistical Feature-Guided Diffusion Model for Sensor-based Human Activity Recognition

A Novel Deep Learning Bi-GRU-I Model for Real-Time Human Activity Recognition Using Inertial Sensors

Semi-Supervised Adversarial Learning Using LSTM for Human Activity Recognition

Timestamp-Supervised Wearable-Based Activity Segmentation and Recognition With Contrastive Learning and Order-Preserving Optimal Transport