pith. sign in

arxiv: 2605.16570 · v1 · pith:J5JQLY7Bnew · submitted 2026-05-15 · 📊 stat.CO · stat.ML

A Cubing Strategy for Identifying Stable Hyperparameter Regions for Uncertainty Quantification in Spatial Deep Learning

Isaac Amouzou , Ben Seiyon Lee This is my paper

Pith reviewed 2026-05-19 21:11 UTC · model grok-4.3

classification 📊 stat.CO stat.ML
keywords spatial deep learningMC dropoutuncertainty quantificationhyperparameter tuningcubing strategypredictive intervalsspatial statistics
0
0 comments X

The pith

A cubing strategy systematically identifies stable hyperparameter regions for well-calibrated MC dropout in spatial deep learning.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper develops a cubing-based framework to partition the hyperparameter space for Monte Carlo dropout models in spatial settings. It scores sub-regions of the space using rules that compare performance to a statistical baseline model serving as calibration anchor. The method is tested in simulations with different spatial dependence and on a large land surface temperature dataset, where it yields predictive intervals that match or exceed the baseline. Sympathetic readers would value this because ad-hoc tuning of dropout and related parameters often leads to poorly calibrated uncertainty in deep spatial models.

Core claim

The central discovery is that recursively cubing the hyperparameter space and evaluating regions relative to a statistical baseline identifies stable areas where MC dropout produces well-calibrated predictive intervals for spatial predictions, performing competitively or better than the baseline in both simulated and real data.

What carries the argument

The cubing-based diagnostic framework, which recursively partitions the hyperparameter space to identify stable regions for MC dropout performance.

If this is right

  • Practitioners can replace manual tuning with this systematic procedure for uncertainty in spatial deep learning.
  • The approach works across multiple regimes of spatial dependence.
  • Large remotely sensed datasets can be handled with reliable uncertainty estimates.
  • Scoring against the baseline provides a concrete way to validate hyperparameter choices.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the cubing works here, it might generalize to hyperparameter stability checks in non-spatial deep learning tasks.
  • Interactions between dropout rate, weight decay, and standard deviation multiplier could be better understood through this partitioning.
  • Extensions could combine this with other uncertainty methods like deep ensembles.

Load-bearing premise

The statistical baseline model serves as a reliable calibration anchor when scoring hyperparameter regions for MC dropout performance.

What would settle it

A simulation or dataset where the cubing-identified regions produce predictive intervals with calibration scores worse than those from the baseline model or standard tuning would contradict the central claim.

Figures

Figures reproduced from arXiv: 2605.16570 by Ben Seiyon Lee, Isaac Amouzou.

Figure 1
Figure 1. Figure 1: Example execution of the cubing algorithm ( [PITH_FULL_IMAGE:figures/full_fig_p017_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Illustration of the neural network architecture for an example with five input features. [PITH_FULL_IMAGE:figures/full_fig_p035_2.png] view at source ↗
read the original abstract

Spatially referenced datasets have become increasingly prevalent across many fields, largely driven by advances in data collection methods such as satellite remote sensing. In many applications, predictions at unobserved locations are accompanied by reliable uncertainty estimates. While deep learning methods provide both scalable and accurate models for spatial predictions, there remains no clear consensus for addressing uncertainty quantification in spatial deep learning. Monte Carlo (MC) dropout has become a popular approach for uncertainty quantification, yet existing implementations typically focus on tuning the dropout rate while fixing other influential hyperparameters, such as weight decay and the predictive standard deviation multiplier, often through ad-hoc or manual tuning. We propose a cubing-based diagnostic framework that recursively partitions the hyperparameter space to identify stable regions where MC dropout yields well-calibrated predictive intervals. The approach evaluates hyperparameter regions using scoring rules relative to a statistical baseline model, which serves as a calibration anchor. Through a simulation study spanning multiple spatial dependence regimes as well as a large remotely-sensed land surface temperature dataset, we demonstrate that our approach produces competitive or superior predictive intervals compared to the baseline model. Our methodology provides practitioners with a systematic procedure for incorporating uncertainty quantification into spatial deep learning models.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript proposes a cubing-based diagnostic framework that recursively partitions the hyperparameter space of MC dropout (including dropout rate, weight decay, and predictive standard deviation multiplier) to identify stable regions yielding well-calibrated predictive intervals for spatial deep learning. Hyperparameter regions are scored using proper scoring rules relative to a statistical baseline model that serves as a calibration anchor. The approach is demonstrated on a simulation study spanning multiple spatial dependence regimes and on a large remotely-sensed land surface temperature dataset, with the central claim that the selected regions produce competitive or superior predictive intervals compared to the baseline.

Significance. If the empirical results hold after quantitative verification, the work supplies a systematic, reproducible procedure for hyperparameter selection in spatial uncertainty quantification, addressing a practical gap where ad-hoc tuning of MC dropout is common. The explicit use of an external statistical baseline as anchor supplies an independent reference point that could help avoid purely internal circularity in deep-learning UQ evaluations.

major comments (2)
  1. Abstract and evaluation sections: the central claim that the cubing procedure yields 'competitive or superior predictive intervals' is stated without any reported quantitative metrics (coverage rates, CRPS values, interval widths, or error bars) or details on how post-hoc region selection was performed. This absence leaves the strength of the evidence for the main result difficult to assess.
  2. Evaluation / simulation study: no quantitative diagnostics (coverage probabilities, PIT histograms, or calibration plots) are supplied for the statistical baseline model itself across the simulated spatial dependence regimes. Without such verification, it remains possible that the cubing procedure simply identifies MC-dropout configurations that reproduce any miscalibration present in the baseline rather than demonstrating genuine improvement.
minor comments (1)
  1. Notation for the cubing recursion and the scoring-rule aggregation across partitions could be clarified with an explicit algorithm box or pseudocode.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed report. We address the two major comments below and will incorporate the suggested additions into the revised manuscript to strengthen the empirical support for our claims.

read point-by-point responses

  1. Referee: Abstract and evaluation sections: the central claim that the cubing procedure yields 'competitive or superior predictive intervals' is stated without any reported quantitative metrics (coverage rates, CRPS values, interval widths, or error bars) or details on how post-hoc region selection was performed. This absence leaves the strength of the evidence for the main result difficult to assess.

    Authors: We agree that the abstract and evaluation sections currently present the central claim without the supporting quantitative metrics. In the revision we will insert a concise summary of the key metrics (coverage rates, CRPS, interval widths, and standard errors) directly into the abstract and will expand the evaluation section with tables and figures that report these quantities for both the simulation study and the land-surface temperature application. We will also add an explicit description of the post-hoc region selection procedure, including the scoring rules, the recursive partitioning thresholds, and the stability criteria used to designate a region as stable. revision: yes

  2. Referee: Evaluation / simulation study: no quantitative diagnostics (coverage probabilities, PIT histograms, or calibration plots) are supplied for the statistical baseline model itself across the simulated spatial dependence regimes. Without such verification, it remains possible that the cubing procedure simply identifies MC-dropout configurations that reproduce any miscalibration present in the baseline rather than demonstrating genuine improvement.

    Authors: The referee is correct that the current manuscript does not supply calibration diagnostics for the statistical baseline. We will add, in the revised simulation-study section, coverage probabilities, PIT histograms, and calibration plots for the baseline model under each spatial dependence regime. These diagnostics will be placed immediately before the comparison with the cubing-selected MC-dropout regions so that readers can directly evaluate whether the selected regions improve upon, rather than merely replicate, the baseline’s calibration properties. revision: yes

Circularity Check

0 steps flagged

No significant circularity; evaluation anchored to external statistical baseline

full rationale

The paper's core procedure evaluates hyperparameter regions for MC dropout via scoring rules relative to a separate statistical baseline model treated as calibration anchor. The reported demonstration of competitive or superior predictive intervals is a direct empirical comparison against this independent anchor across simulations and real data, rather than any quantity that reduces by construction to a fitted parameter or self-referential definition inside the cubing procedure itself. No equations, self-citations, or uniqueness claims are presented that would force the result to equal its inputs. The derivation chain therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that a statistical baseline provides an appropriate calibration reference and that recursive partitioning can isolate stable regions without introducing new fitted constants beyond standard hyperparameters.

axioms (1)
  • domain assumption The statistical baseline model serves as a reliable calibration anchor for evaluating MC dropout hyperparameter regions.
    Invoked when scoring regions and declaring them stable or superior.

pith-pipeline@v0.9.0 · 5735 in / 1149 out tokens · 37116 ms · 2026-05-19T21:11:10.490683+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

294 extracted references · 294 canonical work pages · 14 internal anchors

  1. [1]

    Journal of Machine Learning Research , volume=

    Deep out-of-distribution uncertainty quantification via weight entropy maximization , author=. Journal of Machine Learning Research , volume=

    work page

  2. [2]

    Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    Rate-in: Information-driven adaptive dropout rates for improved inference-time uncertainty estimation , author=. Proceedings of the Computer Vision and Pattern Recognition Conference , pages=

    work page

  3. [3]

    IEEE transactions on medical imaging , volume=

    Confidence calibration and predictive uncertainty estimation for deep medical image segmentation , author=. IEEE transactions on medical imaging , volume=. 2020 , publisher=

    work page 2020

  4. [4]

    nature , volume=

    Learning representations by back-propagating errors , author=. nature , volume=. 1986 , publisher=

    work page 1986

  5. [5]

    stat , volume=

    Pathologies of factorised gaussian and mc dropout posteriors in bayesian neural networks , author=. stat , volume=

    work page

  6. [6]

    International conference on machine learning , pages=

    What are Bayesian neural network posteriors really like? , author=. International conference on machine learning , pages=. 2021 , organization=

    work page 2021

  7. [7]

    arXiv preprint arXiv:1909.13550 , year=

    Well-calibrated model uncertainty with temperature scaling for dropout variational inference , author=. arXiv preprint arXiv:1909.13550 , year=

    work page arXiv 1909

  8. [8]

    arXiv preprint arXiv:2112.05000 , year=

    The peril of popular deep learning uncertainty estimation methods , author=. arXiv preprint arXiv:2112.05000 , year=

    work page arXiv

  9. [9]

    Northern Lights Deep Learning Conference 2026 , year=

    Unreliable Monte Carlo Dropout Uncertainty Estimation , author=. Northern Lights Deep Learning Conference 2026 , year=

    work page 2026

  10. [10]

    Journal of the American Statistical Association , volume=

    A multi-resolution approximation for massive spatial datasets , author=. Journal of the American Statistical Association , volume=. 2017 , publisher=

    work page 2017

  11. [11]

    Advances in Neural Information Processing Systems , volume=

    What uncertainties do we need in bayesian deep learning for computer vision? , author=. Advances in Neural Information Processing Systems , volume=

    work page

  12. [12]

    arXiv preprint arXiv:1805.04829 , year=

    Spatial uncertainty sampling for end-to-end control , author=. arXiv preprint arXiv:1805.04829 , year=

    work page arXiv

  13. [13]

    Artificial Intelligence for the Earth Systems , volume=

    Evaluating probabilistic deep learning methods for uncertainty quantification of precipitation bias correction , author=. Artificial Intelligence for the Earth Systems , volume=. 2025 , publisher=

    work page 2025

  14. [14]

    Applied Sciences , volume=

    Monte Carlo dropout neural networks for forecasting sinusoidal time series: Performance evaluation and uncertainty quantification , author=. Applied Sciences , volume=. 2025 , publisher=

    work page 2025

  15. [15]

    Available at SSRN 4668687 , year=

    An uncertainty estimation model for health signal prediction , author=. Available at SSRN 4668687 , year=

    work page

  16. [16]

    Multitask Learning

    Caruana, Rich. Multitask Learning. Learning to Learn. 1998. doi:10.1007/978-1-4615-5529-2_5

    work page doi:10.1007/978-1-4615-5529-2_5 1998

  17. [17]

    Machine learning , volume=

    A Bayesian/information theoretic model of learning to learn via multiple task sampling , author=. Machine learning , volume=. 1997 , publisher=

    work page 1997

  18. [18]

    An Overview of Multi-Task Learning in Deep Neural Networks

    An overview of multi-task learning in deep neural networks , author=. arXiv preprint arXiv:1706.05098 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  19. [19]

    Proceedings of the Tenth International Conference on Machine Learning , pages=

    Multitask learning: A knowledge-based source of inductive bias1 , author=. Proceedings of the Tenth International Conference on Machine Learning , pages=

    work page

  20. [20]

    1999 , publisher=

    Interpolation of spatial data: some theory for kriging , author=. 1999 , publisher=

    work page 1999

  21. [21]

    Journal of Animal Ecology , volume=

    Spatio-temporal variation in lifelong telomere dynamics in a long-term ecological study , author=. Journal of Animal Ecology , volume=. 2018 , publisher=

    work page 2018

  22. [22]

    BMC infectious diseases , volume=

    Modelling and predicting the spatio-temporal spread of COVID-19 in Italy , author=. BMC infectious diseases , volume=. 2020 , publisher=

    work page 2020

  23. [23]

    Journal of Quantitative Criminology , volume=

    Spatio-temporal interaction of urban crime , author=. Journal of Quantitative Criminology , volume=. 2008 , publisher=

    work page 2008

  24. [24]

    IEEE Transactions on Big Data , volume=

    Discovering congestion propagation patterns in spatio-temporal traffic data , author=. IEEE Transactions on Big Data , volume=. 2016 , publisher=

    work page 2016

  25. [25]

    Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies , volume=

    Detecting urban anomalies using multiple spatio-temporal data sources , author=. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies , volume=. 2018 , publisher=

    work page 2018

  26. [26]

    ACM Computing Surveys (CSUR) , volume=

    Spatio-temporal data mining: A survey of problems and methods , author=. ACM Computing Surveys (CSUR) , volume=. 2018 , publisher=

    work page 2018

  27. [27]

    2011 , publisher=

    Spatio-temporal heterogeneity: Concepts and analyses , author=. 2011 , publisher=

    work page 2011

  28. [28]

    Functional ecology , volume=

    Non-optimal animal movement in human-altered landscapes , author=. Functional ecology , volume=. 2007 , publisher=

    work page 2007

  29. [29]

    Trends in ecology & evolution , volume=

    From birds to butterflies: animal movement patterns and stable isotopes , author=. Trends in ecology & evolution , volume=. 2004 , publisher=

    work page 2004

  30. [30]

    Current Landscape Ecology Reports , volume=

    Hierarchical species distribution models , author=. Current Landscape Ecology Reports , volume=. 2016 , publisher=

    work page 2016

  31. [31]

    Stochastic environmental research and risk assessment , volume=

    Species distribution modeling: a statistical review with focus in spatio-temporal issues , author=. Stochastic environmental research and risk assessment , volume=. 2018 , publisher=

    work page 2018

  32. [32]

    Spatio-temporal variability of rainfall indices and their teleconnections with El Ni

    Gehlot, Lalit Kumar and Jibhakate, Shubham M and Sharma, Priyank J and Patel, PL and Timbadiya, PV , journal=. Spatio-temporal variability of rainfall indices and their teleconnections with El Ni. 2021 , publisher=

    work page 2021

  33. [33]

    Biometrics , volume=

    Effects of residual smoothing on the posterior of the fixed effects in disease-mapping models , author=. Biometrics , volume=. 2006 , publisher=

    work page 2006

  34. [34]

    Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=

    Dimension reduction and alleviation of confounding for spatial generalized linear mixed models , author=. Journal of the Royal Statistical Society: Series B (Statistical Methodology) , volume=. 2013 , publisher=

    work page 2013

  35. [35]

    Journal of the American Statistical Association , volume=

    Classes of nonseparable, spatio-temporal stationary covariance functions , author=. Journal of the American Statistical Association , volume=. 1999 , publisher=

    work page 1999

  36. [36]

    Journal of the American Statistical Association , volume=

    Nonseparable, stationary covariance functions for space--time data , author=. Journal of the American Statistical Association , volume=. 2002 , publisher=

    work page 2002

  37. [37]

    2015 , publisher=

    Statistics for spatio-temporal data , author=. 2015 , publisher=

    work page 2015

  38. [38]

    2019 , publisher=

    Spatio-temporal Statistics with R , author=. 2019 , publisher=

    work page 2019

  39. [39]

    Wiley Interdisciplinary Reviews: Computational Statistics , volume=

    Modern perspectives on statistics for spatio-temporal data , author=. Wiley Interdisciplinary Reviews: Computational Statistics , volume=. 2015 , publisher=

    work page 2015

  40. [40]

    Environmental and Ecological Statistics , volume=

    A hierarchical Bayesian non-linear spatio-temporal model for the spread of invasive species with application to the Eurasian Collared-Dove , author=. Environmental and Ecological Statistics , volume=. 2008 , publisher=

    work page 2008

  41. [41]

    Oikos , pages=

    Allee effect and population dynamics in the Glanville fritillary butterfly , author=. Oikos , pages=. 1998 , publisher=

    work page 1998

  42. [42]

    Dispersal ecology , pages=

    Invasion and the range expansion of species: effects of long-distance dispersal , author=. Dispersal ecology , pages=. 2002 , publisher=

    work page 2002

  43. [43]

    Wavelet Applications in Signal and Image Processing VIII , volume=

    Wavelets and radial basis functions: A unifying perspective , author=. Wavelet Applications in Signal and Image Processing VIII , volume=. 2000 , organization=

    work page 2000

  44. [44]

    2010 International Conference on Logistics Systems and Intelligent Management (ICLSIM) , volume=

    Traffic flow prediction based on wavelet transform and radial basis function network , author=. 2010 International Conference on Logistics Systems and Intelligent Management (ICLSIM) , volume=. 2010 , organization=

    work page 2010

  45. [45]

    Computers in Physics , volume=

    Solving PDEs using wavelets , author=. Computers in Physics , volume=. 1997 , publisher=

    work page 1997

  46. [46]

    Friston et al.(Eds.), Statistical Parametric Mapping: Models for Brain Imaging

    Spatio-temporal models for EEG , author=. Friston et al.(Eds.), Statistical Parametric Mapping: Models for Brain Imaging. Elsevier , year=

    work page

  47. [47]

    Communications on pure and applied mathematics , volume=

    Biorthogonal bases of compactly supported wavelets , author=. Communications on pure and applied mathematics , volume=. 1992 , publisher=

    work page 1992

  48. [48]

    Technometrics , number=

    PICAR: An Efficient Extendable Approach for Fitting Hierarchical Spatial Models , author=. Technometrics , number=. 2021 , publisher=

    work page 2021

  49. [49]

    Journal of Computational and Graphical Statistics , volume=

    Robust Markov chain Monte Carlo methods for spatial generalized linear mixed models , author=. Journal of Computational and Graphical Statistics , volume=. 2006 , publisher=

    work page 2006

  50. [50]

    Test , volume=

    A general science-based framework for dynamical spatio-temporal models , author=. Test , volume=. 2010 , publisher=

    work page 2010

  51. [51]

    2005 , publisher=

    Gaussian Markov random fields: theory and applications , author=. 2005 , publisher=

    work page 2005

  52. [52]

    Journal of statistical planning and inference , volume=

    A close look at the spatial structure implied by the CAR and SAR models , author=. Journal of statistical planning and inference , volume=. 2004 , publisher=

    work page 2004

  53. [53]

    Journal of the American Statistical Association , volume=

    Basis function models for animal movement , author=. Journal of the American Statistical Association , volume=. 2017 , publisher=

    work page 2017

  54. [54]

    Animal Behaviour , volume=

    Simulation of the mate-finding behaviour of pine shoot beetles, Tomicus piniperda , author=. Animal Behaviour , volume=. 1991 , publisher=

    work page 1991

  55. [55]

    Methods in Ecology and Evolution , volume=

    Joint modelling of multi-scale animal movement data using hierarchical hidden Markov models , author=. Methods in Ecology and Evolution , volume=. 2019 , publisher=

    work page 2019

  56. [56]

    Handbook of Markov Chain Monte Carlo , pages=

    Gaussian random field models for spatial data , author=. Handbook of Markov Chain Monte Carlo , pages=. 2011 , publisher=

    work page 2011

  57. [57]

    Geostatistics Wollongong , volume=

    Spatio-temporal kriging of soil water content , author=. Geostatistics Wollongong , volume=

    work page

  58. [58]

    Statistica Neerlandica , volume=

    Aspects of second-order analysis of structured inhomogeneous spatio-temporal point processes , author=. Statistica Neerlandica , volume=. 2012 , publisher=

    work page 2012

  59. [59]

    Stochastic environmental research and risk assessment , volume=

    Strict positive definiteness in geostatistics , author=. Stochastic environmental research and risk assessment , volume=. 2018 , publisher=

    work page 2018

  60. [60]

    International Journal of Electrical Power & Energy Systems , volume=

    Wind power scenario generation with non-separable spatio-temporal covariance function and fluctuation-based clustering , author=. International Journal of Electrical Power & Energy Systems , volume=. 2021 , publisher=

    work page 2021

  61. [61]

    arXiv preprint arXiv:2110.10249 , year=

    Neural Stochastic Partial Differential Equations , author=. arXiv preprint arXiv:2110.10249 , year=

    work page arXiv

  62. [62]

    Spatial statistics , volume=

    Dynamic spatio-temporal models for spatial data , author=. Spatial statistics , volume=. 2017 , publisher=

    work page 2017

  63. [63]

    Journal of Agricultural, Biological and Environmental Statistics , volume=

    Agent-based inference for animal movement and selection , author=. Journal of Agricultural, Biological and Environmental Statistics , volume=. 2010 , publisher=

    work page 2010

  64. [64]

    Transactions of The Royal Society of Tropical Medicine and Hygiene , year=

    A comparison of modelling the spatio-temporal pattern of disease: a case study of schistosomiasis japonica in Anhui Province, China , author=. Transactions of The Royal Society of Tropical Medicine and Hygiene , year=

    work page

  65. [65]

    Stochastic environmental research and risk assessment , volume=

    Monthly stream flow forecasting via dynamic spatio-temporal models , author=. Stochastic environmental research and risk assessment , volume=. 2015 , publisher=

    work page 2015

  66. [66]

    Environmetrics: The official journal of the International Environmetrics Society , volume=

    Testing for separability of space--time covariances , author=. Environmetrics: The official journal of the International Environmetrics Society , volume=. 2005 , publisher=

    work page 2005

  67. [67]

    Scandinavian Journal of Statistics , volume=

    A class of convolution-based models for spatio-temporal processes with Non-separable covariance structure , author=. Scandinavian Journal of Statistics , volume=. 2010 , publisher=

    work page 2010

  68. [68]

    Proceedings of the Royal Society A , volume=

    Locally stationary spatio-temporal interpolation of Argo profiling float data , author=. Proceedings of the Royal Society A , volume=. 2018 , publisher=

    work page 2018

  69. [69]

    Wiley interdisciplinary reviews: computational statistics , volume=

    Spatio-temporal processes , author=. Wiley interdisciplinary reviews: computational statistics , volume=. 2010 , publisher=

    work page 2010

  70. [70]

    Journal of Agricultural, Biological and Environmental Statistics , volume=

    Hierarchical nonlinear spatio-temporal agent-based models for collective animal movement , author=. Journal of Agricultural, Biological and Environmental Statistics , volume=. 2017 , publisher=

    work page 2017

  71. [71]

    arXiv preprint arXiv:2009.04003 , year=

    Bayesian Inverse Reinforcement Learning for Collective Animal Movement , author=. arXiv preprint arXiv:2009.04003 , year=

    work page arXiv 2009

  72. [72]

    Spatial Statistics , volume=

    Animal movement models with mechanistic selection functions , author=. Spatial Statistics , volume=. 2020 , publisher=

    work page 2020

  73. [73]

    Methods in Ecology and Evolution , volume=

    Animal movement models for migratory individuals and groups , author=. Methods in Ecology and Evolution , volume=. 2018 , publisher=

    work page 2018

  74. [74]

    Scientific reports , volume=

    Both nearest neighbours and long-term affiliates predict individual locations during collective movement in wild baboons , author=. Scientific reports , volume=. 2016 , publisher=

    work page 2016

  75. [75]

    Biostatistics , volume=

    On the change of support problem for spatio-temporal data , author=. Biostatistics , volume=. 2001 , publisher=

    work page 2001

  76. [76]

    Cartography and Geographic Information Science , volume=

    Design and evaluation of a geovisual analytics system for uncovering patterns in spatio-temporal event data , author=. Cartography and Geographic Information Science , volume=. 2017 , publisher=

    work page 2017

  77. [77]

    Data mining and knowledge discovery handbook , pages=

    Spatio-temporal clustering , author=. Data mining and knowledge discovery handbook , pages=. 2009 , publisher=

    work page 2009

  78. [78]

    Proceedings of the national academy of sciences , volume=

    Cluster failure: Why fMRI inferences for spatial extent have inflated false-positive rates , author=. Proceedings of the national academy of sciences , volume=. 2016 , publisher=

    work page 2016

  79. [79]

    Proceedings of the 1st ACM SIGSPATIAL international workshop on analytics for big geospatial data , pages=

    Spatiotemporal data mining in the era of big spatial data: algorithms and applications , author=. Proceedings of the 1st ACM SIGSPATIAL international workshop on analytics for big geospatial data , pages=

    work page

  80. [80]

    Soil Science Society of America Journal , volume=

    Soil property and class maps of the conterminous United States at 100-meter spatial resolution , author=. Soil Science Society of America Journal , volume=. 2018 , publisher=

    work page 2018

Showing first 80 references.