Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral models

<p>Uncertainty quantification is a crucial step in the practical application of soil spectral models, particularly in supporting real-world decision making and risk assessment. While machine learning has made remarkable strides in predicting various physiochemical properties of soils using spe...

Full description

Saved in:
Bibliographic Details
Main Authors: Y.-C. Huang, J. Padarian, B. Minasny, A. B. McBratney
Format: Article
Language:English
Published: Copernicus Publications 2025-07-01
Series:SOIL
Online Access:https://soil.copernicus.org/articles/11/553/2025/soil-11-553-2025.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849424642141323264
author Y.-C. Huang
J. Padarian
B. Minasny
A. B. McBratney
author_facet Y.-C. Huang
J. Padarian
B. Minasny
A. B. McBratney
author_sort Y.-C. Huang
collection DOAJ
description <p>Uncertainty quantification is a crucial step in the practical application of soil spectral models, particularly in supporting real-world decision making and risk assessment. While machine learning has made remarkable strides in predicting various physiochemical properties of soils using spectroscopy, its practical utility in decision making remains limited without quantified uncertainty. Despite its importance, uncertainty quantification is rarely incorporated into soil spectral models, with existing methods facing significant limitations. Existing methods are either computationally demanding, fail to achieve the desired coverage of observed data, or struggle to handle out-of-domain uncertainty. This study introduces an innovative application of Monte Carlo conformal prediction (MC-CP) to quantify uncertainty in deep-learning models for predicting clay content from mid-infrared spectroscopy. We compared MC-CP with two established methods: (1) Monte Carlo dropout and (2) conformal prediction. Monte Carlo dropout generates prediction intervals for each sample and can address larger uncertainties associated with out-of-domain data. Conformal prediction, on the other hand, guarantees ideal coverage of true values but generates unnecessarily wide prediction intervals, making it overly conservative for many practical applications. Using 39 177 samples from the mid-infrared spectral library of the Kellogg Soil Survey Laboratory to build convolutional neural networks, we found that Monte Carlo dropout itself falls short in achieving the desired coverage – its 90 % prediction intervals only covered the observed values in 74 % of the cases, well below the expected 90 % coverage. In contrast, MC-CP successfully combines the strengths of both methods. It achieved a prediction interval coverage probability of 91 %, closely matching the expected 90 % coverage and far surpassing the performance of the Monte Carlo dropout. Additionally, the mean prediction interval width for MC-CP was 9.05 %, narrower than the conformal prediction's 11.11 %. The success of MC-CP enhances the real-world applicability of soil spectral models, paving the way for their integration into large-scale machine learning models, such as soil inference systems, and further transforming decision making and risk assessment in soil science.</p>
format Article
id doaj-art-e658ad809f074a2bb0ddcacb594bbc17
institution Kabale University
issn 2199-3971
2199-398X
language English
publishDate 2025-07-01
publisher Copernicus Publications
record_format Article
series SOIL
spelling doaj-art-e658ad809f074a2bb0ddcacb594bbc172025-08-20T03:30:05ZengCopernicus PublicationsSOIL2199-39712199-398X2025-07-011155356310.5194/soil-11-553-2025Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral modelsY.-C. Huang0J. Padarian1B. Minasny2A. B. McBratney3School of Life and Environmental Science & Sydney Institute of Agriculture, The University of Sydney, Sydney, NSW, AustraliaSchool of Life and Environmental Science & Sydney Institute of Agriculture, The University of Sydney, Sydney, NSW, AustraliaSchool of Life and Environmental Science & Sydney Institute of Agriculture, The University of Sydney, Sydney, NSW, AustraliaSchool of Life and Environmental Science & Sydney Institute of Agriculture, The University of Sydney, Sydney, NSW, Australia<p>Uncertainty quantification is a crucial step in the practical application of soil spectral models, particularly in supporting real-world decision making and risk assessment. While machine learning has made remarkable strides in predicting various physiochemical properties of soils using spectroscopy, its practical utility in decision making remains limited without quantified uncertainty. Despite its importance, uncertainty quantification is rarely incorporated into soil spectral models, with existing methods facing significant limitations. Existing methods are either computationally demanding, fail to achieve the desired coverage of observed data, or struggle to handle out-of-domain uncertainty. This study introduces an innovative application of Monte Carlo conformal prediction (MC-CP) to quantify uncertainty in deep-learning models for predicting clay content from mid-infrared spectroscopy. We compared MC-CP with two established methods: (1) Monte Carlo dropout and (2) conformal prediction. Monte Carlo dropout generates prediction intervals for each sample and can address larger uncertainties associated with out-of-domain data. Conformal prediction, on the other hand, guarantees ideal coverage of true values but generates unnecessarily wide prediction intervals, making it overly conservative for many practical applications. Using 39 177 samples from the mid-infrared spectral library of the Kellogg Soil Survey Laboratory to build convolutional neural networks, we found that Monte Carlo dropout itself falls short in achieving the desired coverage – its 90 % prediction intervals only covered the observed values in 74 % of the cases, well below the expected 90 % coverage. In contrast, MC-CP successfully combines the strengths of both methods. It achieved a prediction interval coverage probability of 91 %, closely matching the expected 90 % coverage and far surpassing the performance of the Monte Carlo dropout. Additionally, the mean prediction interval width for MC-CP was 9.05 %, narrower than the conformal prediction's 11.11 %. The success of MC-CP enhances the real-world applicability of soil spectral models, paving the way for their integration into large-scale machine learning models, such as soil inference systems, and further transforming decision making and risk assessment in soil science.</p>https://soil.copernicus.org/articles/11/553/2025/soil-11-553-2025.pdf
spellingShingle Y.-C. Huang
J. Padarian
B. Minasny
A. B. McBratney
Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral models
SOIL
title Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral models
title_full Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral models
title_fullStr Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral models
title_full_unstemmed Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral models
title_short Using Monte Carlo conformal prediction to evaluate the uncertainty of deep-learning soil spectral models
title_sort using monte carlo conformal prediction to evaluate the uncertainty of deep learning soil spectral models
url https://soil.copernicus.org/articles/11/553/2025/soil-11-553-2025.pdf
work_keys_str_mv AT ychuang usingmontecarloconformalpredictiontoevaluatetheuncertaintyofdeeplearningsoilspectralmodels
AT jpadarian usingmontecarloconformalpredictiontoevaluatetheuncertaintyofdeeplearningsoilspectralmodels
AT bminasny usingmontecarloconformalpredictiontoevaluatetheuncertaintyofdeeplearningsoilspectralmodels
AT abmcbratney usingmontecarloconformalpredictiontoevaluatetheuncertaintyofdeeplearningsoilspectralmodels