More Like this

1. Flexible and Fast Spatial Return Level Estimation Via a Spatially Fused Penalty

   https://doi.org/10.1080/10618600.2021.1938584  

   Sass, Danielle; Li, Bo; Reich, Brian J. (July 2021, Journal of Computational and Graphical Statistics)

   null (Ed.)

   Spatial extremes are common for climate data as the observations are usually referenced by geographic locations and dependent when they are nearby. An important goal of extremes modeling is to estimate the T-year return level. Among the methods suitable for modeling spatial extremes, perhaps the simplest and fastest approach is the spatial generalized extreme value (GEV) distribution and the spatial generalized Pareto distribution (GPD) that assume marginal independence and only account for dependence through the parameters. Despite the simplicity, simulations have shown that return level estimation using the spatial GEV and spatial GPD still provides satisfactory results compared to max-stable processes, which are asymptotically justified models capable of representing spatial dependence among extremes. However, the linear functions used to model the spatially varying coefficients are restrictive and may be violated.We propose a flexible and fast approach based on the spatial GEV and spatial GPD by introducing fused lasso and fused ridge penalty for parameter regularization. This enables improved return level estimation for large spatial extremes compared to the existing methods. Supplemental files for this article are available online. 

   more » « less
2. Beyond Point Prediction: Capturing Zero-Inflated & Heavy-Tailed Spatiotemporal Data with Deep Extreme Mixture Models

   https://doi.org/10.1145/3534678.3539464  

   Wilson, Tyler; McDonald, Andrew; Galib, Asadullah Hill; Tan, Pang-Ning; Luo, Lifeng (August 2022, KDD '22: Proceedings of the 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining)

   Zhang, Aidong; Rangwala, Huzefa (Ed.)

   Zero-inflated, heavy-tailed spatiotemporal data is common across science and engineering, from climate science to meteorology and seismology. A central modeling objective in such settings is to forecast the intensity, frequency, and timing of extreme and non-extreme events; yet in the context of deep learning, this objective presents several key challenges. First, a deep learning framework applied to such data must unify a mixture of distributions characterizing the zero events, moderate events, and extreme events. Second, the framework must be capable of enforcing parameter constraints across each component of the mixture distribution. Finally, the framework must be flexible enough to accommodate for any changes in the threshold used to define an extreme event after training. To address these challenges, we propose Deep Extreme Mixture Model (DEMM), fusing a deep learning-based hurdle model with extreme value theory to enable point and distribution prediction of zero-inflated, heavy-tailed spatiotemporal variables. The framework enables users to dynamically set a threshold for defining extreme events at inference-time without the need for retraining. We present an extensive experimental analysis applying DEMM to precipitation forecasting, and observe significant improvements in point and distribution prediction. All code is available at https://github.com/andrewmcdonald27/DeepExtremeMixtureModel. 

   more » « less
3. DeepExtrema: A Deep Learning Approach for Forecasting Block Maxima in Time Series Data

   https://doi.org/10.24963/ijcai.2022/413  

   Galib, Asadullah Hill; McDonald, Andrew; Wilson, Tyler; Luo, Lifeng; Tan, Pang-Ning (July 2022, Proceedings of the Thirty-First International Joint Conference on Artificial Intelligence)

   Accurate forecasting of extreme values in time series is critical due to the significant impact of extreme events on human and natural systems. This paper presents DeepExtrema, a novel framework that combines a deep neural network (DNN) with generalized extreme value (GEV) distribution to forecast the block maximum value of a time series. Implementing such a network is a challenge as the framework must preserve the inter-dependent constraints among the GEV model parameters even when the DNN is initialized. We describe our approach to address this challenge and present an architecture that enables both conditional mean and quantile prediction of the block maxima. The extensive experiments performed on both real-world and synthetic data demonstrated the superiority of DeepExtrema compared to other baseline methods. 

   more » « less
4. Self-Recover: Forecasting Block Maxima in Time Series from Predictors with Disparate Temporal Coverage Using Self-Supervised Learning

   https://doi.org/10.24963/ijcai.2023/414  

   Galib, Asadullah Hill; McDonald, Andrew; Tan, Pang-Ning; Luo, Lifeng (August 2023, Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence (IJCAI 2023))

   Forecasting the block maxima of a future time window is a challenging task due to the difficulty in inferring the tail distribution of a target variable. As the historical observations alone may not be sufficient to train robust models to predict the block maxima, domain-driven process models are often available in many scientific domains to supplement the observation data and improve the forecast accuracy. Unfortunately, coupling the historical observations with process model outputs is a challenge due to their disparate temporal coverage. This paper presents Self-Recover, a deep learning framework to predict the block maxima of a time window by employing self-supervised learning to address the varying temporal data coverage problem. Specifically Self-Recover uses a combination of contrastive and generative self-supervised learning schemes along with a denoising autoencoder to impute the missing values. The framework also combines representations of the historical observations with process model outputs via a residual learning approach and learns the generalized extreme value (GEV) distribution characterizing the block maxima values. This enables the framework to reliably estimate the block maxima of each time window along with its confidence interval. Extensive experiments on real-world datasets demonstrate the superiority of Self-Recover compared to other state-of-the-art forecasting methods. 

   more » « less
5. Global analysis of temporal clusters of storm surges

   https://doi.org/10.1017/cft.2025.10008  

   Martín, Ariadna; Jane, Robert; Enriquez, Alejandra R; Wahl, Thomas (January 2025, Cambridge Prisms: Coastal Futures)

   Abstract Temporal storm surge clustering refers to a series of events affecting the same region within a short period of time, which can strongly influence coastal flooding impacts and erosion. Here, we analyze global storm surge clustering from tide gauges and a state-of-the-art global model hindcast to identify geographical hotspots of extreme storm surge clusters and assess event frequencies. We study the spatial distribution as well as the contribution of different event intensities to clustering. On average, globally, 92% of coastal locations show significant temporal clustering for 1-year return period events, and 25% for 5-year return level events, although notable spatial differences exist. Our results reveal two distinct clustering regimes: (i) short timescale clustering, where events occur in rapid succession (intra-annual), and (ii) long timescales (inter-annual), providing varying recovery times between events. We also test the validity of assuming a Poisson distribution, commonly used in storm surge frequency analyses. Our results show that >80% of the stations analyzed do not follow a Poisson distribution, at least when including events that are not the most extreme but exceeded, for example, the 1-year return level. These findings offer insights into temporal clustering dynamics of storm surges and their implications for coastal hazard assessments. 

   more » « less