Reconstruction of incomplete wildfire data using deep generative models

DOI10.1007/S10687-022-00459-1arXiv2201.06153OpenAlexW4317211522WikidataQ117820525 ScholiaQ117820525MaRDI QIDQ6100554FDOQ6100554

Tomislav Ivek, Domagoj Vlah

Publication date: 12 May 2023

Published in: Extremes (Search for Journal in Brave)

Abstract: We present our submission to the Extreme Value Analysis 2021 Data Challenge in which teams were asked to accurately predict distributions of wildfire frequency and size within spatio-temporal regions of missing data. For the purpose of this competition we developed a variant of the powerful variational autoencoder models dubbed the Conditional Missing data Importance-Weighted Autoencoder (CMIWAE). Our deep latent variable generative model requires little to no feature engineering and does not necessarily rely on the specifics of scoring in the Data Challenge. It is fully trained on incomplete data, with the single objective to maximize log-likelihood of the observed wildfire information. We mitigate the effects of the relatively low number of training samples by stochastic sampling from a variational latent variable distribution, as well as by ensembling a set of CMIWAE models trained and validated on different splits of the provided data. The presented approach is not domain-specific and is amenable to application in other missing data recovery tasks with tabular or image-like information conditioned on auxiliary information.

Full work available at URL: https://arxiv.org/abs/2201.06153

zbMATH Keywords

deep learningensemblewildfiresdata reconstructionconvolutional neural networkvariational autoencoderextreme value analysis conference challenge

Mathematics Subject Classification ID

Statistical distribution theory (62Exx)

Cites Work

• Title not available (Why is that?)
• Simple statistical gradient-following algorithms for connectionist reinforcement learning

Cited In (3)

• A marginal modelling approach for predicting wildfire extremes across the contiguous United States
• Data-driven surrogate model with latent data assimilation: application to wildfire forecasting
• Ranks of elliptic curves and deep neural networks

This page was built for publication: Reconstruction of incomplete wildfire data using deep generative models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6100554)