{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/extremeweather-a-large-scale-climate-dataset","title":"ExtremeWeather: A large-scale climate dataset for semi-supervised detection, localization, and understanding of extreme weather events","arxiv_id":"1612.02095","date":"2016-12-07","proceeding":"NeurIPS 2017 12","authors":["Evan Racah","Christopher Beckham","Tegan Maharaj","Samira Ebrahimi Kahou","Prabhat","Christopher Pal"],"abstract":"Then detection and identification of extreme weather events in large-scale\nclimate simulations is an important problem for risk management, informing\ngovernmental policy decisions and advancing our basic understanding of the\nclimate system. Recent work has shown that fully supervised convolutional\nneural networks (CNNs) can yield acceptable accuracy for classifying well-known\ntypes of extreme weather events when large amounts of labeled data are\navailable. However, many different types of spatially localized climate\npatterns are of interest including hurricanes, extra-tropical cyclones, weather\nfronts, and blocking events among others. Existing labeled data for these\npatterns can be incomplete in various ways, such as covering only certain years\nor geographic areas and having false negatives. This type of climate data\ntherefore poses a number of interesting machine learning challenges. We present\na multichannel spatiotemporal CNN architecture for semi-supervised bounding box\nprediction and exploratory data analysis. We demonstrate that our approach is\nable to leverage temporal information and unlabeled data to improve the\nlocalization of extreme weather events. Further, we explore the representations\nlearned by our model in order to better understand this important data. We\npresent a dataset, ExtremeWeather, to encourage machine learning research in\nthis area and to help facilitate further work in understanding and mitigating\nthe effects of climate change. The dataset is available at\nextremeweatherdataset.github.io and the code is available at\nhttps://github.com/eracah/hur-detect.","url_abs":"http://arxiv.org/abs/1612.02095v2","url_pdf":"http://arxiv.org/pdf/1612.02095v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"extremeweather-a-large-scale-climate-dataset","repo_url":"https://github.com/eracah/hur-detect","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"machine-learning","task_name":"BIG-bench Machine Learning"},{"task_slug":"blocking","task_name":"Blocking"},{"task_slug":"management","task_name":"Management"}],"methods":[],"datasets_introduced":[{"slug":"extremeweather","name":"ExtremeWeather","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1612.02095","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}