CoChat

Personalizing Activity Recognition Models Through Quantifying Different Types of Uncertainty Using Wearable Sensors

Akbari, Ali, Jafari, Roozbeh · 2020 · IEEE Transactions on Biomedical Engineering
Cited by 50

The DOI matches its Crossref record; 12 of the 12 references it cites were found in a registry or an index.

Abstract

Recognizing activities of daily living (ADL) provides vital contextual information that enhances the effectiveness of various mobile health and wellness applications. Development of wearable motion sensors along with machine learning algorithms offer a great opportunity for ADL recognition. However, the performance of the ADL recognition systems may significantly degrade when they are used by a new user due to inter-subject variability. This issue limits the usability of these systems. In this paper, we propose a deep learning assisted personalization framework for ADL recognition with the aim to maximize the personalization performance while minimizing solicitation of inputs or labels from the user to reduce user's burden. The proposed framework consists of unsupervised retraining of automatic feature extraction layers and supervised fine-tuning of classification layers through a novel active learning model based on a given model's uncertainty. We design a Bayesian deep convolutional neural network with stochastic latent variables that allows us to estimate both aleatoric (data-dependent) and epistemic (model-dependent) uncertainties in recognition task. In this study, for the first time, we show how distinguishing between the two aforementioned sources of uncertainty leads to more effective active learning. The experimental results show that our proposed method improves the accuracy of ADL recognition on a new user by 25% on average compared to the case of using a model for a new user with no personalization with an average final accuracy of 89.2%. Moreover, our method achieves higher personalization accuracy while significantly reducing user's burden in terms of soliciting inputs and labels compared to other methods.

Overview

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/tbme.2019.2963816

Cites

12 of 12 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Citations

Cites

Show all 12 references

Research works printed but not linked: 36

  1. weiss, 0, The impact of personalization on smartphone-based activity recognition, Proc Workshops at 26th AAAI Conf Artificial Intell
  2. 10.1145/3267305.3267528 · DOI · Paper page
  3. 10.1145/3302506.3310391 · DOI · Paper page
  4. 10.1109/JIOT.2018.2856119 · DOI
  5. 10.1109/ICSENS.2016.7808590 · DOI · Paper page
  6. yang, 0, Deep convolutional neural networks on multichannel time series for human activity recognition, Proc 24th Int Joint Conf Artif Intell, 3995
  7. zhao, 0, Cross-people mobile-phone based activity recognition, Proc 22nd Int Joint Conf Artif Intell, 2545
  8. zhao, 2011, Human action recognition based on non-linear svm decision tree, J Comput Inform Syst, 7, 2461
  9. 10.1145/3267305.3267525 · DOI
  10. 10.1145/2638728.2641674 · DOI
  11. avci, 0, Activity recognition using inertial sensing for healthcare, wellbeing and sports applications: A survey, Proc 23th Int Conf Architecture Comput Syst, 1
  12. 10.1007/978-3-540-70994-7_36 · DOI
  13. 10.1109/ISWCS.2010.5624490 · DOI
  14. 10.1109/JSEN.2016.2562599 · DOI
  15. 10.1109/ASPDAC.2008.4483992 · DOI
  16. 10.1145/3055004.3055015 · DOI
  17. 10.4108/icst.pervasivehealth.2012.248600 · DOI
  18. 10.1145/2971648.2971691 · DOI
  19. 10.1145/2971648.2971701 · DOI · Paper page
  20. 10.1109/MNET.2016.7474340 · DOI · Paper page
  21. 10.1109/JBHI.2018.2869779 · DOI
  22. 10.1145/3214269 · DOI · Paper page
  23. añazco, 0, Smoking activity recognition using a single wrist IMU and deep learning light, Proc 2nd Int Conf Digit Signal Process, 48
  24. 10.1109/ICHI.2016.100 · DOI
  25. 10.1145/3302505.3310087 · DOI
  26. 10.1109/SAS.2017.7894068 · DOI
  27. 10.3414/ME0592 · DOI · Paper page
  28. 10.1016/j.eswa.2012.09.004 · DOI · Paper page
  29. erhan, 2010, Why does unsupervised pre-training help deep learning?, J Mach Learn Res, 11, 625
  30. 10.1007/978-3-642-02085-8_11 · DOI
  31. 10.1109/BHI.2019.8834505 · DOI
  32. 10.1007/978-3-319-46493-0_36 · DOI · Paper page
  33. vincent, 2010, Stacked denoising autoencoders: Learning useful representations in a deep network with a local denoising criterion, J Mach Learn Res, 11, 3371
  34. diethe, 0, Active transfer learning for activity recognition, Proc Eur Symp Artif Neural Netw Comput Intell Mach Learn, 429
  35. sabato, 2016, Interactive algorithms: From pool to stream, 29th Annual Conference on Learning Theory, 49, 23
  36. 10.1109/PERCOM.2017.7917864 · DOI · Paper page

Cited by

Metadata

Authors
Akbari, A., & Jafari, R.
Venue
IEEE Transactions on Biomedical Engineering
Published
2020-09
Volume
67
Issue
9
Pages
2530-2541
ISSN
  • 0018-9294 (ISSN-L)
  • 1558-2531
OpenAlex
W2997626223
Cited by
50
References
12
FWCI
3.41
Type
journal article
Language
en
Open access
closed
Funders
  • National Science Foundation (CNS-1734039, ECCS-1509063)
  • National Institutes of Health (1R01EB028106-01)

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.