Lifelong learning in sensor-based human activity recognition

Juan Ye; Simon Andrew Dobson; Franco Zambonelli

doi:10.1109/MPRV.2019.2913933

Lifelong learning in sensor-based human activity recognition

Juan Ye^*, Simon Andrew Dobson^*, Franco Zambonelli

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

Abstract

Human activity recognition (HAR) systems will be increasingly deployed in real-world environments and for long periods of time. This significantly challenges current approaches to HAR, which have to account for changes in activity routines, the evolution of situations, and sensing technologies. Driven by these challenges, in this paper, we argue the need to move beyond learning to lifelong machine learning—with the ability to incrementally and continuously adapt to changes in the environment being learned. We introduce a conceptual framework for lifelong machine learning to structure various relevant proposals in the area and identify some key research challenges that remain.

Original language	English
Pages (from-to)	49-58
Number of pages	10
Journal	IEEE Pervasive Computing
Volume	18
Issue number	3
DOIs	https://doi.org/10.1109/MPRV.2019.2913933
Publication status	Published - 18 Nov 2019

Access to Document

10.1109/MPRV.2019.2913933

PC-2018-04-0038.v4
Copyright © 2019 IEEE Computer Society. This work has been made available online in accordance with publisher policies or with permission. Permission for further reuse of this content should be sought from the publisher or the rights holder. This is the author created accepted manuscript following peer review and may differ slightly from the final published version. The final published version of this work is available at https://doi.org/10.1109/MPRV.2019.2913933
Accepted author manuscript, 2.33 MB

Cite this

@article{1f432c26c49a4674b877f0f5a9a6c07f,

title = "Lifelong learning in sensor-based human activity recognition",

abstract = "Human activity recognition (HAR) systems will be increasingly deployed in real-world environments and for long periods of time. This significantly challenges current approaches to HAR, which have to account for changes in activity routines, the evolution of situations, and sensing technologies. Driven by these challenges, in this paper, we argue the need to move beyond learning to lifelong machine learning—with the ability to incrementally and continuously adapt to changes in the environment being learned. We introduce a conceptual framework for lifelong machine learning to structure various relevant proposals in the area and identify some key research challenges that remain.",

author = "Juan Ye and Dobson, {Simon Andrew} and Franco Zambonelli",

year = "2019",

month = nov,

day = "18",

doi = "10.1109/MPRV.2019.2913933",

language = "English",

volume = "18",

pages = "49--58",

journal = "IEEE Pervasive Computing",

issn = "1536-1268",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "3",

}

TY - JOUR

T1 - Lifelong learning in sensor-based human activity recognition

AU - Ye, Juan

AU - Dobson, Simon Andrew

AU - Zambonelli, Franco

PY - 2019/11/18

Y1 - 2019/11/18

N2 - Human activity recognition (HAR) systems will be increasingly deployed in real-world environments and for long periods of time. This significantly challenges current approaches to HAR, which have to account for changes in activity routines, the evolution of situations, and sensing technologies. Driven by these challenges, in this paper, we argue the need to move beyond learning to lifelong machine learning—with the ability to incrementally and continuously adapt to changes in the environment being learned. We introduce a conceptual framework for lifelong machine learning to structure various relevant proposals in the area and identify some key research challenges that remain.

AB - Human activity recognition (HAR) systems will be increasingly deployed in real-world environments and for long periods of time. This significantly challenges current approaches to HAR, which have to account for changes in activity routines, the evolution of situations, and sensing technologies. Driven by these challenges, in this paper, we argue the need to move beyond learning to lifelong machine learning—with the ability to incrementally and continuously adapt to changes in the environment being learned. We introduce a conceptual framework for lifelong machine learning to structure various relevant proposals in the area and identify some key research challenges that remain.

U2 - 10.1109/MPRV.2019.2913933

DO - 10.1109/MPRV.2019.2913933

M3 - Article

SN - 1536-1268

VL - 18

SP - 49

EP - 58

JO - IEEE Pervasive Computing

JF - IEEE Pervasive Computing

IS - 3

ER -

Lifelong learning in sensor-based human activity recognition

Abstract

Access to Document

Fingerprint

Cite this