Home > Research > Publications & Outputs > Adaptive policies for perimeter surveillance pr...

Research

Associated organisational units

Links

Text available via DOI:

Keywords

View graph of relations

Adaptive policies for perimeter surveillance problems

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Adaptive policies for perimeter surveillance problems. / Grant, James A.; Leslie, David S.; Glazebrook, Kevin et al.
In: European Journal of Operational Research, Vol. 283, No. 1, 16.05.2020, p. 265-278.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Grant, JA, Leslie, DS, Glazebrook, K, Szechtman, R & Letchford, A 2020, 'Adaptive policies for perimeter surveillance problems', European Journal of Operational Research, vol. 283, no. 1, pp. 265-278. https://doi.org/10.1016/j.ejor.2019.11.004

APA

Grant, J. A., Leslie, D. S., Glazebrook, K., Szechtman, R., & Letchford, A. (2020). Adaptive policies for perimeter surveillance problems. European Journal of Operational Research, 283(1), 265-278. https://doi.org/10.1016/j.ejor.2019.11.004

Vancouver

Grant JA, Leslie DS, Glazebrook K, Szechtman R, Letchford A. Adaptive policies for perimeter surveillance problems. European Journal of Operational Research. 2020 May 16;283(1):265-278. Epub 2019 Nov 6. doi: 10.1016/j.ejor.2019.11.004

Author

Grant, James A. ; Leslie, David S. ; Glazebrook, Kevin et al. / Adaptive policies for perimeter surveillance problems. In: European Journal of Operational Research. 2020 ; Vol. 283, No. 1. pp. 265-278.

Bibtex

@article{d3750b528ed74bd9a0049b3c44fbcb81,
title = "Adaptive policies for perimeter surveillance problems",
abstract = "We consider the problem of sequentially choosing observation regions along a line, with an aim of maximising the detection of events of interest. Such a problem may arise when monitoring the movements of endangered or migratory species, detecting crossings of a border, policing activities at sea, and in many other settings. In each case, the key operational challenge is to learn an allocation of surveillance resources which maximises successful detection of events of interest. We present a combinatorial multi-armed bandit model with Poisson rewards and a novel filtered feedback mechanism - arising from the failure to detect certain intrusions - where reward distributions are dependent on the actions selected. Our solution method is an upper confidence bound approach and we derive upper and lower bounds on its expected performance. We prove that the gap between these bounds is of constant order, and demonstrate empirically that our approach is more reliable in simulated problems than competing algorithms.",
keywords = "cs.LG, stat.ML, Applied probability, Stochastic processes, Uncertainty modelling, OR in defence",
author = "Grant, {James A.} and Leslie, {David S.} and Kevin Glazebrook and Roberto Szechtman and Adam Letchford",
year = "2020",
month = may,
day = "16",
doi = "10.1016/j.ejor.2019.11.004",
language = "English",
volume = "283",
pages = "265--278",
journal = "European Journal of Operational Research",
issn = "0377-2217",
publisher = "Elsevier Science B.V.",
number = "1",

}

RIS

TY - JOUR

T1 - Adaptive policies for perimeter surveillance problems

AU - Grant, James A.

AU - Leslie, David S.

AU - Glazebrook, Kevin

AU - Szechtman, Roberto

AU - Letchford, Adam

PY - 2020/5/16

Y1 - 2020/5/16

N2 - We consider the problem of sequentially choosing observation regions along a line, with an aim of maximising the detection of events of interest. Such a problem may arise when monitoring the movements of endangered or migratory species, detecting crossings of a border, policing activities at sea, and in many other settings. In each case, the key operational challenge is to learn an allocation of surveillance resources which maximises successful detection of events of interest. We present a combinatorial multi-armed bandit model with Poisson rewards and a novel filtered feedback mechanism - arising from the failure to detect certain intrusions - where reward distributions are dependent on the actions selected. Our solution method is an upper confidence bound approach and we derive upper and lower bounds on its expected performance. We prove that the gap between these bounds is of constant order, and demonstrate empirically that our approach is more reliable in simulated problems than competing algorithms.

AB - We consider the problem of sequentially choosing observation regions along a line, with an aim of maximising the detection of events of interest. Such a problem may arise when monitoring the movements of endangered or migratory species, detecting crossings of a border, policing activities at sea, and in many other settings. In each case, the key operational challenge is to learn an allocation of surveillance resources which maximises successful detection of events of interest. We present a combinatorial multi-armed bandit model with Poisson rewards and a novel filtered feedback mechanism - arising from the failure to detect certain intrusions - where reward distributions are dependent on the actions selected. Our solution method is an upper confidence bound approach and we derive upper and lower bounds on its expected performance. We prove that the gap between these bounds is of constant order, and demonstrate empirically that our approach is more reliable in simulated problems than competing algorithms.

KW - cs.LG

KW - stat.ML

KW - Applied probability

KW - Stochastic processes

KW - Uncertainty modelling

KW - OR in defence

U2 - 10.1016/j.ejor.2019.11.004

DO - 10.1016/j.ejor.2019.11.004

M3 - Journal article

VL - 283

SP - 265

EP - 278

JO - European Journal of Operational Research

JF - European Journal of Operational Research

SN - 0377-2217

IS - 1

ER -