The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling
Agent-based models return spatiotemporal information used to process time series of specific parameters for specific individuals called “agents”. For complex, advanced and detailed models, this typically comes at the expense of high computing times and requires access to important computing resource...
Published in: | Journal of Marine Science and Engineering |
---|---|
Main Authors: | , , , |
Format: | Text |
Language: | English |
Published: |
Multidisciplinary Digital Publishing Institute
2022
|
Subjects: | |
Online Access: | https://doi.org/10.3390/jmse10070899 |
id |
ftmdpi:oai:mdpi.com:/2077-1312/10/7/899/ |
---|---|
record_format |
openpolar |
spelling |
ftmdpi:oai:mdpi.com:/2077-1312/10/7/899/ 2023-08-20T04:05:34+02:00 The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling Dominic Lagrois Tyler R. Bonnell Ankita Shukla Clément Chion agris 2022-06-29 application/pdf https://doi.org/10.3390/jmse10070899 EN eng Multidisciplinary Digital Publishing Institute Physical Oceanography https://dx.doi.org/10.3390/jmse10070899 https://creativecommons.org/licenses/by/4.0/ Journal of Marine Science and Engineering; Volume 10; Issue 7; Pages: 899 agent-based models machine learning numerical solutions underwater acoustic techniques Text 2022 ftmdpi https://doi.org/10.3390/jmse10070899 2023-08-01T05:32:12Z Agent-based models return spatiotemporal information used to process time series of specific parameters for specific individuals called “agents”. For complex, advanced and detailed models, this typically comes at the expense of high computing times and requires access to important computing resources. This paper provides an example on how machine learning and artificial intelligence can help predict an agent-based model’s output values at regular intervals without having to rely on time-consuming numerical calculations. Gradient-boosting XGBoost under GNU package’s R was used in the social-ecological agent-based model 3MTSim to interpolate, in the time domain, sound pressure levels received at the agents’ positions that were occupied by the endangered St. Lawrence Estuary and Saguenay Fjord belugas and caused by anthropomorphic noise of nearby transiting merchant vessels. A mean error of 3.23 ± 3.76(1σ) dB on received sound pressure levels was predicted when compared to ground truth values that were processed using rigorous, although time-consuming, numerical algorithms. The computing time gain was significant, i.e., it was estimated to be 10-fold higher than the ground truth simulation, whilst maintaining the original temporal resolution. Text Beluga* MDPI Open Access Publishing Journal of Marine Science and Engineering 10 7 899 |
institution |
Open Polar |
collection |
MDPI Open Access Publishing |
op_collection_id |
ftmdpi |
language |
English |
topic |
agent-based models machine learning numerical solutions underwater acoustic techniques |
spellingShingle |
agent-based models machine learning numerical solutions underwater acoustic techniques Dominic Lagrois Tyler R. Bonnell Ankita Shukla Clément Chion The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling |
topic_facet |
agent-based models machine learning numerical solutions underwater acoustic techniques |
description |
Agent-based models return spatiotemporal information used to process time series of specific parameters for specific individuals called “agents”. For complex, advanced and detailed models, this typically comes at the expense of high computing times and requires access to important computing resources. This paper provides an example on how machine learning and artificial intelligence can help predict an agent-based model’s output values at regular intervals without having to rely on time-consuming numerical calculations. Gradient-boosting XGBoost under GNU package’s R was used in the social-ecological agent-based model 3MTSim to interpolate, in the time domain, sound pressure levels received at the agents’ positions that were occupied by the endangered St. Lawrence Estuary and Saguenay Fjord belugas and caused by anthropomorphic noise of nearby transiting merchant vessels. A mean error of 3.23 ± 3.76(1σ) dB on received sound pressure levels was predicted when compared to ground truth values that were processed using rigorous, although time-consuming, numerical algorithms. The computing time gain was significant, i.e., it was estimated to be 10-fold higher than the ground truth simulation, whilst maintaining the original temporal resolution. |
format |
Text |
author |
Dominic Lagrois Tyler R. Bonnell Ankita Shukla Clément Chion |
author_facet |
Dominic Lagrois Tyler R. Bonnell Ankita Shukla Clément Chion |
author_sort |
Dominic Lagrois |
title |
The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling |
title_short |
The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling |
title_full |
The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling |
title_fullStr |
The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling |
title_full_unstemmed |
The Gradient-Boosting Method for Tackling High Computing Demand in Underwater Acoustic Propagation Modeling |
title_sort |
gradient-boosting method for tackling high computing demand in underwater acoustic propagation modeling |
publisher |
Multidisciplinary Digital Publishing Institute |
publishDate |
2022 |
url |
https://doi.org/10.3390/jmse10070899 |
op_coverage |
agris |
genre |
Beluga* |
genre_facet |
Beluga* |
op_source |
Journal of Marine Science and Engineering; Volume 10; Issue 7; Pages: 899 |
op_relation |
Physical Oceanography https://dx.doi.org/10.3390/jmse10070899 |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.3390/jmse10070899 |
container_title |
Journal of Marine Science and Engineering |
container_volume |
10 |
container_issue |
7 |
container_start_page |
899 |
_version_ |
1774716124157444096 |