Showing 1–2 of 2 results for author: Ssematimba, J

Modelling calibration uncertainty in networks of environmental sensors

Authors: Michael Thomas Smith, Magnus Ross, Joel Ssematimba, Pablo A. Alvarado, Mauricio Alvarez, Engineer Bainomugisha, Richard Wilkinson

Abstract: Networks of low-cost sensors are becoming ubiquitous, but often suffer from poor accuracies and drift. Regular colocation with reference sensors allows recalibration but is complicated and expensive. Alternatively the calibration can be transferred using low-cost, mobile sensors. However inferring the calibration (with uncertainty) becomes difficult. We propose a variational approach to model the… ▽ More Networks of low-cost sensors are becoming ubiquitous, but often suffer from poor accuracies and drift. Regular colocation with reference sensors allows recalibration but is complicated and expensive. Alternatively the calibration can be transferred using low-cost, mobile sensors. However inferring the calibration (with uncertainty) becomes difficult. We propose a variational approach to model the calibration across the network. We demonstrate the approach on synthetic and real air pollution data, and find it can perform better than the state of the art (multi-hop calibration). We extend it to categorical data produced by citizen-scientist labelling. In Summary: The method achieves uncertainty-quantified calibration, which has been one of the barriers to low-cost sensor deployment and citizen-science research. △ Less

Submitted 9 May, 2022; v1 submitted 4 May, 2022; originally announced May 2022.

Comments: 31 pages (23 pages of content, 4 pages of references, 4 supplementary). 11 figures. 4 tables. Submitted to Journal of the Royal Statistical Society. Series C

MSC Class: 60G15

Machine Learning for a Low-cost Air Pollution Network

Authors: Michael T. Smith, Joel Ssematimba, Mauricio A. Alvarez, Engineer Bainomugisha

Abstract: Data collection in economically constrained countries often necessitates using approximate and biased measurements due to the low-cost of the sensors used. This leads to potentially invalid predictions and poor policies or decision making. This is especially an issue if methods from resource-rich regions are applied without handling these additional constraints. In this paper we show, through the… ▽ More Data collection in economically constrained countries often necessitates using approximate and biased measurements due to the low-cost of the sensors used. This leads to potentially invalid predictions and poor policies or decision making. This is especially an issue if methods from resource-rich regions are applied without handling these additional constraints. In this paper we show, through the use of an air pollution network example, how using probabilistic machine learning can mitigate some of the technical constraints. Specifically we experiment with modelling the calibration for individual sensors as either distributions or Gaussian processes over time, and discuss the wider issues around the decision process. △ Less

Submitted 28 November, 2019; originally announced November 2019.

Comments: Presented at NeurIPS 2019 Workshop on Machine Learning for the Developing World

MSC Class: 60G15