Evaluation Procedures for Forecasting with Spatio-Temporal Data

Oliveira, Mariana; Torgo, Luı́s; Costa, Vítor Santos

doi:10.1007/978-3-030-10925-7_43

Cited by 14 publications

(17 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When working with spatiotemporal data, the separation between training and validation becomes tricky. The spatial and temporal dimensions in the dataset cannot be ignored and strongly affect the independence between training and validation data (Roberts et al, 2017;Oliveira et al, 2019). Depending on how the cross validation is performed, the obtained performance will be indicative of one of these two di- Figure 11.…”

Section: Training Deep Learning Models With Spatiotemporal Datamentioning

confidence: 99%

“…Nonetheless, more recent developments in the field of machine learning and optimization enabled the use of deeper network structures than the threelayer ANN of Steiner et al (2005). These deeper ANNs, which remain unexploited in glaciology, allow us to capture more nonlinear structures in the data even for relatively small datasets (Ingrassia and Morlini, 2005;Olson et al, 2018).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning applied to glacier evolution modelling

et al. 2020

View full text Add to dashboard Cite

Abstract. We present a novel approach to simulate and reconstruct annual glacier-wide surface mass balance (SMB) series based on a deep artificial neural network (ANN; i.e. deep learning). This method has been included as the SMB component of an open-source regional glacier evolution model. While most glacier models tend to incorporate more and more physical processes, here we take an alternative approach by creating a parameterized model based on data science. Annual glacier-wide SMBs can be simulated from topo-climatic predictors using either deep learning or Lasso (least absolute shrinkage and selection operator; regularized multilinear regression), whereas the glacier geometry is updated using a glacier-specific parameterization. We compare and cross-validate our nonlinear deep learning SMB model against other standard linear statistical methods on a dataset of 32 French Alpine glaciers. Deep learning is found to outperform linear methods, with improved explained variance (up to +64 % in space and +108 % in time) and accuracy (up to +47 % in space and +58 % in time), resulting in an estimated r2 of 0.77 and a root-mean-square error (RMSE) of 0.51 m w.e. Substantial nonlinear structures are captured by deep learning, with around 35 % of nonlinear behaviour in the temporal dimension. For the glacier geometry evolution, the main uncertainties come from the ice thickness data used to initialize the model. These results should encourage the use of deep learning in glacier modelling as a powerful nonlinear tool, capable of capturing the nonlinearities of the climate and glacier systems, that can serve to reconstruct or simulate SMB time series for individual glaciers in a whole region for past and future climates.

show abstract

Section: Training Deep Learning Models With Spatiotemporal Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Deep learning applied to glacier evolution modelling

et al. 2020

View full text Add to dashboard Cite

show abstract

“…When working with spatiotemporal data, the separation between training and validation becomes tricky. The spatial and temporal dimensions in the dataset cannot be ignored, and strongly affect the independence between training and validation data (Roberts et al, 2017;Oliveira et al, 2019). Depending on how the cross-validation is performed, the obtained performance will be indicative of one of these two dimensions.…”

Section: Training Deep Learning Models With Spatiotemporal Datamentioning

confidence: 99%

Deep learning applied to glacier evolution modelling

Bolíbar¹,

Rabatel²,

Gouttevin³

et al. 2019

Preprint

View full text Add to dashboard Cite

Abstract. We present a parameterized glacier evolution model, with a surface mass balance (SMB) component based on a deep artificial neural network (i.e. deep learning). While most glacier models tend to incorporate more and more physical processes, here we take an alternative approach by creating a parameterized model based on data science. Annual glacier-wide SMBs can be simulated using either deep learning or Lasso (regularized multilinear regression), whereas the glacier geometry is updated using a glacier-specific parameterization. We compare and cross-validate our nonlinear deep learning SMB model against other standard linear statistical methods on a dataset of 32 French alpine glaciers. Deep learning is found to outperform linear methods, with improved explained variance (up to +64 % in space and +108 % in time) and accuracy (up to +47 % in space and +58 % in time), resulting in an estimated r2 of 0.77 and RMSE of 0.51 m.w.e. Substantial nonlinear structures are captured by deep learning, with around 35 % of nonlinear behaviour in the temporal dimension. For the glacier geometry evolution, the main uncertainties come from the ice thickness data used to initialize the model. These results should encourage the use of deep learning in glacier modelling as a powerful nonlinear tool, capable of capturing the nonlinearities of the climate and glacier systems, that can serve to reconstruct or simulate SMB time series for individual glaciers at a regional scale for past and future climates.

show abstract

“…Therefore, we design several ML and ML ensemble models using blocked sequential procedure (Cerqueira et al, 2017;Oliveira et al, 2019) to generate out-of-bag predictions and evaluate their performance when forecasting corn yields. In addition, we investigate the effect of having complete or partial in-season weather knowledge when forecasting yields.…”

Section: Introductionmentioning

confidence: 99%

Forecasting Corn Yield With Machine Learning Ensembles

2020

View full text Add to dashboard Cite

The emergence of new technologies to synthesize and analyze big data with highperformance computing has increased our capacity to more accurately predict crop yields. Recent research has shown that machine learning (ML) can provide reasonable predictions faster and with higher flexibility compared to simulation crop modeling. However, a single machine learning model can be outperformed by a "committee" of models (machine learning ensembles) that can reduce prediction bias, variance, or both and is able to better capture the underlying distribution of the data. Yet, there are many aspects to be investigated with regard to prediction accuracy, time of the prediction, and scale. The earlier the prediction during the growing season the better, but this has not been thoroughly investigated as previous studies considered all data available to predict yields. This paper provides a machine leaning based framework to forecast corn yields in three US Corn Belt states (Illinois, Indiana, and Iowa) considering complete and partial inseason weather knowledge. Several ensemble models are designed using blocked sequential procedure to generate out-of-bag predictions. The forecasts are made in county-level scale and aggregated for agricultural district and state level scales. Results show that the proposed optimized weighted ensemble and the average ensemble are the most precise models with RRMSE of 9.5%. Stacked LASSO makes the least biased predictions (MBE of 53 kg/ha), while other ensemble models also outperformed the base learners in terms of bias. On the contrary, although random k-fold cross-validation is replaced by blocked sequential procedure, it is shown that stacked ensembles perform not as good as weighted ensemble models for time series data sets as they require the data to be non-IID to perform favorably. Comparing our proposed model forecasts with the literature demonstrates the acceptable performance of forecasts made by our proposed ensemble model. Results from the scenario of having partial in-season weather knowledge reveals that decent yield forecasts with RRMSE of 9.2% can be made as early as June 1 st. Moreover, it was shown that the proposed model performed better than individual models and benchmark ensembles at agricultural district and statelevel scales as well as county-level scale. To find the marginal effect of each input feature on the forecasts made by the proposed ensemble model, a methodology is suggested that is the basis for finding feature importance for the ensemble model. The findings

show abstract

Evaluation Procedures for Forecasting with Spatio-Temporal Data

Cited by 14 publications

References 31 publications

Deep learning applied to glacier evolution modelling

Deep learning applied to glacier evolution modelling

Deep learning applied to glacier evolution modelling

Forecasting Corn Yield With Machine Learning Ensembles

Contact Info

Product

Resources

About