Summary of Deep Learning Joint Extremes Of Metocean Variables Using the Spar Model, by Ed Mackay et al.
Deep learning joint extremes of metocean variables using the SPAR model
by Ed Mackay, Callum Murphy-Barltrop, Jordan Richards, Philip Jonathan
First submitted to arxiv on: 20 Dec 2024
Categories
- Main: Machine Learning (stat.ML)
- Secondary: Machine Learning (cs.LG); Methodology (stat.ME)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary A novel deep learning framework is introduced for estimating multivariate joint extremes of metocean variables, utilizing the Semi-Parametric Angular-Radial (SPAR) model. By transforming the problem into one of modelling an angular density and the tail of a univariate radial variable conditioned on angle, the SPAR approach models the tail of the radial variable using a generalized Pareto distribution, extending univariate extreme value theory to multivariate settings. The method is applied in higher dimensions, with a case study involving five metocean variables: wind speed, direction, wave height, period, and direction. Empirical modelling is used for the angular variable, while GP model parameters are approximated using fully-connected deep neural networks. This data-driven approach offers flexibility in dependence structures, computationally efficient training routines, and fewer assumptions about underlying distributions compared to existing methods. Diagnostic plots demonstrate that fitted models provide a good description of joint extremes. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to predict extreme values for multiple ocean-related factors like wind speed, direction, wave height, period, and direction. It uses a special type of artificial intelligence called deep learning to make these predictions. The approach is flexible and can be used with many different types of data, making it useful for scientists studying the oceans. |
Keywords
» Artificial intelligence » Deep learning
