Evaluation of Transfer Learning for Human Activity Recognition Among Different Datasets

Human activity recognition is a potential area of research. For better performance, it requires significant amount of labelled data. Collecting labeled activity data is expensive and time-consuming. To solve this problem, transfer learning has been demonstrated very effective as it gathers knowledge from labeled train data of source domain and transfers that knowledge to target domain, which has little or no labeled data. In this paper, we propose unsupervised transfer learning from source dataset to target dataset, which are completely different in terms of number of users and samples. We have used Maximum Mean Discrepancy (MMD) based transfer learning model and compared with base Convolutional Neural Network (CNN) model. We have used 4 datasets for experiment. We have trained the model on a source dataset and then transferred the model to a target dataset, which has no labels to classify activities. We have found that transfer learning model has achieved better performance compared to the base model.

[1]  Zicheng Liu,et al.  Cross-dataset action detection , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[2]  Michael I. Jordan,et al.  Deep Transfer Learning with Joint Adaptation Networks , 2016, ICML.

[3]  Trevor Darrell,et al.  Deep Domain Confusion: Maximizing for Domain Invariance , 2014, CVPR 2014.

[4]  Yiqiang Chen,et al.  Cross-People Mobile-Phone Based Activity Recognition , 2011, IJCAI.

[5]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[6]  Archan Misra,et al.  Scaling Human Activity Recognition via Deep Learning-based Domain Adaptation , 2018, 2018 IEEE International Conference on Pervasive Computing and Communications (PerCom).

[7]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[8]  Yiqiang Chen,et al.  Deep Transfer Learning for Cross-domain Activity Recognition , 2018, ICCSE'18.

[9]  Jin-Hyuk Hong,et al.  Toward Personalized Activity Recognition Systems With a Semipopulation Approach , 2016, IEEE Transactions on Human-Machine Systems.

[10]  Bernt Schiele,et al.  A tutorial on human activity recognition using body-worn inertial sensors , 2014, CSUR.

[11]  Philip S. Yu,et al.  Stratified Transfer Learning for Cross-domain Activity Recognition , 2017, 2018 IEEE International Conference on Pervasive Computing and Communications (PerCom).

[12]  Bernhard Schölkopf,et al.  A Kernel Two-Sample Test , 2012, J. Mach. Learn. Res..

[13]  Thomas Plötz,et al.  Ensembles of Deep LSTM Learners for Activity Recognition using Wearables , 2017, Proc. ACM Interact. Mob. Wearable Ubiquitous Technol..

[14]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[15]  Miguel A. Labrador,et al.  A Survey on Human Activity Recognition using Wearable Sensors , 2013, IEEE Communications Surveys & Tutorials.

[16]  Ivor W. Tsang,et al.  Domain Adaptation via Transfer Component Analysis , 2009, IEEE Transactions on Neural Networks.

[17]  Xiaohui Peng,et al.  Deep Learning for Sensor-based Activity Recognition: A Survey , 2017, Pattern Recognit. Lett..

[18]  Daniel Roggen,et al.  Deep convolutional feature transfer across mobile activity recognition domains, sensor modalities and locations , 2016, SEMWEB.

[19]  H. Shimodaira,et al.  Improving predictive inference under covariate shift by weighting the log-likelihood function , 2000 .

[20]  Takeshi Nishida,et al.  Deep recurrent neural network for mobile human activity recognition with high throughput , 2017, Artificial Life and Robotics.