Ambient Assisted Living (AAL) systems are increasingly being deployed in real-world environments and for long periods of time. This significantly challenges current approaches that require substantial setup investment and cannot account for frequent, unpredictable changes in human behaviours, health conditions, and sensor deployments. The state-of-the-art methodology in studying human activity recognition is cultivated from short-term lab or testbed experimentation, i.e., relying on well-annotated sensor data and assuming no change in activity models. This paper propose a technique, EMILEA, to evolve an activity model over time with new types of activities. This technique novelly integrates two recent advances in continual learning: Net2Net – expanding the architecture of a model while transferring the knowledge from the previous model to the new model and Gradient Episodic Memory – controlling the update on the model parameters to maintain the performance on recognising previously learnt activities. This technique has been evaluated on two real-world, third-party, datasets and demonstrated promising results on enhancing the learning capacity to accommodate new activities that are incrementally introduced to the model while not compromising the accuracy on old activities.
R.Aljundi, P.Chakravarty and T.Tuytelaars, Expert gate: Lifelong learning with a network of experts, in: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 7120–7129. doi:10.1109/CVPR.2017.753.
2.
K.Altun and B.Barshan, Human activity recognition using inertial/magnetic sensor units, in: Proceedings of the First International Conference on Human Behavior Understanding, HBU’10, Springer-Verlag, Berlin, 2010, pp. 38–51. ISBN 978-3-642-14714-2. doi:10.1007/978-3-642-14715-9_5.
3.
K.Altun, B.Barshan and O.Tunçel, Comparative study on classifying human activities with miniature inertial and magnetic sensors, Pattern Recognition43(10) (2010), 3605–3620. doi:10.1016/j.patcog.2010.04.019.
4.
B.Barshan and M.C.Yüksek, Recognizing daily and sports activities in two open source machine learning environments using body-worn sensor units, The Computer Journal57(11) (2014), 1649–1667. doi:10.1093/comjnl/bxt075.
5.
Y.Bengio, J.Louradour, R.Collobert and J.Weston, Curriculum learning, in: Proceedings of the 26th Annual International Conference on Machine Learning, ICML ’09, ACM, New York, NY, 2009, pp. 41–48. ISBN 978-1-60558-516-1. doi:10.1145/1553374.1553380.
A.Chaudhry, M.Ranzato, M.Rohrbach and M.Elhoseiny, Efficient lifelong learning with A-GEM, in: International Conference on Learning Representations, 2019, https://openreview.net/forum?id=Hkf2_sC5FX.
8.
T.Chen, I.Goodfellow and J.Shlens, Net2Net: Accelerating learning via knowledge transfer, 2015, https://arxiv.org/abs/1511.05641.
9.
Z.Chen and B.Liu, Lifelong Machine Learning, Morgan et Claypool Publishers, 2018.
10.
H.T.Cheng, F.T.Sun, M.Griss, P.Davis, J.Li and D.You, Nuactiv: Recognizing unseen new activities using semantic attribute-based learning, in: Proceeding of MobiSys ’13, 2013, pp. 361–374. doi:10.1145/2462456.2464438.
11.
J.L.Elman, Learning and development in neural networks: The importance of starting small, Cognition48(1) (1993), 71–99. doi:10.1016/0010-0277(93)90058-4.
12.
L.Fang, J.Ye and S.Dobson, Discovery and recognition of emerging human activities using a hierarchical mixture of directional statistical models, IEEE Transactions on Knowledge and Data Engineering32(7) (2019), 1304–1316. doi:10.1109/TKDE.2019.2905207.
13.
H.Gjoreski and D.Roggen, Unsupervised online activity discovery using temporal behaviour assumption, in: Proceedings of ISWC ’17, 2017, pp. 42–49. ISBN 978-1-4503-5188-1. doi:10.1145/3123021.3123044.
14.
Z.Gong, P.Zhong and W.Hu, Diversity in machine learning, IEEE Access7 (2019), 64323–64350. doi:10.1109/access.2019.2917620.
15.
R.Kemker, A.Abitino, M.McClure and C.Kanan, Measuring catastrophic forgetting in neural networks, 2017, http://arxiv.org/abs/1708.02072.
Y.Lecun, L.Bottou, Y.Bengio and P.Haffner, Gradient-based learning applied to document recognition, Proceedings of the IEEE86(11) (1998), 2278–2324. doi:10.1109/5.726791.
18.
Z.Li and D.Hoiem, Learning without forgetting, 2016, http://arxiv.org/abs/1606.09282.
19.
D.Lopez-Paz and M.Ranzato, Gradient episodic memory for continuum learning, 2017, http://arxiv.org/abs/1706.08840.
20.
G.Parisi, R.Kemker, J.L.Part, C.Kanan and S.Wermter, Continual lifelong learning with neural networks: A review, Neural Networks113 (2018), 54–71. doi:10.1016/j.neunet.2019.01.012.
21.
G.I.Parisi, J.Tani, C.Weber and S.Wermter, Lifelong learning of human actions with deep neural network self-organization, Neural Networks96 (2017), 137–149. doi:10.1016/j.neunet.2017.09.001.
22.
P.Rashidi and A.Mihailidis, A survey on ambient-assisted living tools for older adults, IEEE Journal of Biomedical and Health Informatics17(3) (2013), 579–590. doi:10.1109/JBHI.2012.2234129.
23.
S.Rebuffi, A.Kolesnikov and C.H.Lampert, iCaRL: Incremental classifier and representation learning, 2016, http://arxiv.org/abs/1611.07725.
24.
A.Reiss and D.Stricker, Introducing a new benchmarked dataset for activity monitoring, in: 2012 16th International Symposium on Wearable Computers, 2012, pp. 108–109. doi:10.1109/ISWC.2012.13.
J.H.Shin, B.Lee and K.S.Park, Detection of abnormal living patterns for elderly living alone using support vector data description, IEEE Transactions on Information Technology in Biomedicine15(3) (2011), 438–448. doi:10.1109/TITB.2011.2113352.
27.
A.Tapus, M.J.Mataric and B.Scassellati, Socially assistive robotics [Grand Challenges of Robotics], IEEE Robotics Automation Magazine14(1) (2007), 35–42. doi:10.1109/MRA.2007.339605.
28.
G.M.van de Ven and A.S.Tolias, Three scenarios for continual learning, 2019, http://arxiv.org/abs/1904.07734.
29.
J.Wang, Y.Chen, L.Hu, X.Peng and P.S.Yu, Stratified transfer learning for cross-domain activity recognition, in: 2018 IEEE International Conference on Pervasive Computing and Communications (PerCom), IEEE, 2018. ISBN 9781538632246. doi:10.1109/percom.2018.8444572.
30.
D.L.Weeks, G.L.Sprint, V.Stilwill, A.L.Meisen-Vehrs and D.J.Cook, Implementing wearable sensors for continuous assessment of daytime heart rate response in inpatient rehabilitation, Telemedicine and e-Health24(12) (2018), 1014–1020. doi:10.1089/tmj.2017.0306.
31.
J.Ye, L.Fang and S.Dobson, Discovery and recognition of unknown activities, in: Proceedings of Ubicomp ’16 Adjunct, ACM, 2016, pp. 783–792. doi:10.1145/2968219.2968288.
32.
J.Ye, F.Zambonelli and S.Dobson, Lifelong learning in sensor-based human activity recognition, in: IEEE Pervasive Computing, 2019. doi:10.1109/PERCOMW.2019.8730783.
33.
J.Yoon, E.Yang, J.Lee and S.J.Hwang, Lifelong learning with dynamically expandable networks, 2017, https://arxiv.org/abs/1708.01547.
