Abstract
Extreme and rare events, such as abnormal spikes in air pollution or weather conditions can have serious repercussions. Many of these sorts of events develop from spatio-temporal processes, and accurate predictions are a most valuable tool in addressing their impact, in a timely manner. In this paper, we propose a new set of resampling strategies for imbalanced spatiotemporal forecasting tasks, by introducing bias into formerly random processes. This spatio-temporal bias includes a hyperparameter that regulates the relative importance of the temporal and spatial dimensions in the selection of observations during under- or over-sampling. We test and compare our proposals against standard versions of the strategies on 10 different georeferenced numeric time series, using 3 distinct off-the-shelf learning algorithms. Experimental results show that our proposal provides an advantage over random resampling strategies in imbalanced spatio-temporal forecasting tasks. Additionally, we also find that valuing an observation's recency is more useful when over-sampling; while valuing its spatial distance to other cases with extreme values is more beneficial when under-sampling.

Abstract
The increasing use of sensor networks has led to an ever larger number of available spatiotemporal datasets. Forecasting applications using this type of data are frequently motivated by important domains such as environmental monitoring. Being able to properly assess the performance of different forecasting approaches is fundamental to achieve progress. However, traditional performance estimation procedures, such as cross-validation, face challenges due to the implicit dependence between observations in spatiotemporal datasets. In this paper, we empirically compare several variants of cross-validation (CV) and out-of-sample (OOS) performance estimation procedures, using both artificially generated and real-world spatiotemporal datasets. Our results show both CV and OOS reporting useful estimates, but they suggest that blocking data in space and/or in time may be useful in mitigating CV's bias to underestimate error. Overall, our study shows the importance of considering data dependencies when estimating the performance of spatiotemporal forecasting models.

Abstract
Extreme and rare events, such as spikes in air pollution or abnormal weather conditions, can have serious repercussions. Many of these sorts of events develop through spatio-temporal processes. Timely and accurate predictions are a most valuable tool in addressing their impact. We propose a new set of resampling strategies for imbalanced spatio-temporal forecasting tasks, which introduce bias into formerly random processes. This bias is a combination of a spatial and a temporal weight, which can be either static or relevance-aware, and includes a hyper-parameter that regulates the relative importance of the temporal and spatial dimensions in the selection of observations during under- or over-sampling. We test and compare our proposals against standard versions of the strategies on 10 different geo-referenced numeric time series, using 3 distinct off-the-shelf learning algorithms. Experimental results show that our proposals provide an advantage over random resampling strategies in imbalanced numerical spatio-temporal forecasting tasks.