[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122307-en":3,"doc-seo-122307-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},122307,687197207057,"Sage","https://ap-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0",8,"Research & Report","ALAS - Active Learning for Autoconversion Rates Prediction from Satellite Data","High-resolution climate simulations, such as ICON-LEM, reveal detailed aerosol–cloud–precipitation interactions but remain prohibitively expensive, limiting their use in space and time. ALAS introduces an active learning framework that queries an oracle simulation using abundant unlabeled satellite-derived data to predict autoconversion rates, a key step in precipitation formation. The method reduces labeled-instance requirements through custom fusion query strategies (WiFi, MeFi) and SHAP-based active feature selection.","ALAS: Active Learning for Autoconversion Rates Prediction from  \nSatellite Data  \nMaria Carolina Novitasari  \nUniversity College London  \nJohannes Quaas  \nUniversit¨at Leipzig, ScaDS.AI  \nMiguel R. D. Rodrigues  \nUniversity College London  \nAbstract  \nHigh-resolution simulations, such as the ICOsahedral Non-hydrostatic Large-Eddy Model (ICON-LEM), provide valuable insights into the complex interactions among aerosols, clouds, and precipitation, which are the major contributors to climate change uncertainty. However, due to their exorbitant computational costs, they can only be employed for a limited period and geographical area. To address this, we propose a more cost-effective method powered by an emerging machine learning approach to better understand the intricate dynamics of the climate system. Our approach involves active learning techniques by leveraging high-resolution climate simulation as an oracle that is queried based on an abundant amount of unlabeled data drawn from satellite observations. In particular, we aim to predict autoconversion rates, a crucial step in precipitation formation, while significantly reducing the need fora large number of labeled instances. In this study, we present novel methods: custom fusion query strategies for labeling instances – weight fusion (WiFi) and merge fusion (MeFi)  \n– along with active feature selection based on SHAP. These methods are designed to tackle real-world challenges – in this case, climate change, with a specific focus on the prediction of autoconversion rates – due to their simplicity and practicality in application.  \n1 INTRODUCTION  \nPrecipitation is a crucial weather and climate phenomenon, with its formation rate being influenced by  \nProceedings of the 27th International Conference on Artificial Intelligence and Statistics (AISTATS) 2024, Valencia, Spain. PMLR: Volume 238 . Copyright 2024 by the author(s) .  \nvarious factors, including interactions among aerosols, clouds, and precipitation itself. Understanding these interactions is vital for improving future climate projections, as they represent a major source of uncertainty in estimating climate change’s radiative forcing (IPCC, 2021) .  \nA prevalent method for investigating intricate interactions within the Earth’s system, such as the interplay between aerosols, clouds, and precipitation, involves the utilization of climate models. These models employ numerical solutions to tackle the differential equations governing the fluid dynamics of the atmosphere and ocean, albeit on a discrete grid. However, they are incapable of representing processes smaller than the grid scale (IPCC, 2021) . Alternatively, recent advancements in computational capabilities have paved the way for high-resolution models with finer grid cells, allowing a more accurate portrayal of small-scale atmospheric processes within a realistic large-area context (e.g., Stevens et al. (2020)) . While these high-resolution models excel in capturing small-scale phenomena, their practical application is often constrained to specific spatial regions and short timeframes due to the significant computational complexity involved.  \nFor instance, the ICOsahedral Non-hydrostatic LargeEddy Model (ICON-LEM) (Z¨angl et al., 2015; Dipankaret al., 2015; Heinze et al., 2017), featuring a horizontal grid resolution of up to around 150 meters, serves asa high-resolution simulation model suitable for simulating small-scale atmospheric processes. However, it is computationally very expensive. For instance, running ICON-LEM to simulate a single hour of climate data over Germany requires around 13 hours on 300 computer nodes and incurs a cost of approximately EUR 100,000 per day (Costa-Sur´os et al., 2020) . Given these high costs, it is imperative to seek alternative approaches for understanding complex climate system.  \nThus, we propose developing a machine learning (ML) model with active learning (AL) techniques to predict autoconversion rates, a key process in p","cbCaiuf1gt6emqWT","https://ap.wps.com/l/cbCaiuf1gt6emqWT","pdf",6372744,1,16,"English","en",105,"# Abstract\n# 1 Introduction","[{\"question\":\"Why is autoconversion rate prediction important in this work?\",\"answer\":\"Autoconversion rate is a crucial process in precipitation formation, directly tied to how cloud droplets grow and transform into raindrops.\"},{\"question\":\"What problem does ALAS address compared with high-resolution simulations like ICON-LEM?\",\"answer\":\"High-resolution simulations are extremely computationally expensive, restricting their practical use to limited regions and short periods; ALAS aims to reduce overall cost and labeled-data needs.\"},{\"question\":\"How does ALAS reduce labeling requirements during training?\",\"answer\":\"It uses active learning where unlabeled satellite data are used to query the oracle simulation, combined with active feature selection based on SHAP and fusion query strategies (WiFi, MeFi).\"}]","ALAS - Active Learning for Autoconversion Rates Prediction from Satellite Data | PDF",1785809927,40,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"alas-active-learning-for-autoconversion-rates-prediction-from-satellite-data","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/alas-active-learning-for-autoconversion-rates-prediction-from-satellite-data/122307/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-04",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why is autoconversion rate prediction important in this work?","Question",{"text":76,"@type":77},"Autoconversion rate is a crucial process in precipitation formation, directly tied to how cloud droplets grow and transform into raindrops.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What problem does ALAS address compared with high-resolution simulations like ICON-LEM?",{"text":81,"@type":77},"High-resolution simulations are extremely computationally expensive, restricting their practical use to limited regions and short periods; ALAS aims to reduce overall cost and labeled-data needs.",{"name":83,"@type":74,"acceptedAnswer":84},"How does ALAS reduce labeling requirements during training?",{"text":85,"@type":77},"It uses active learning where unlabeled satellite data are used to query the oracle simulation, combined with active feature selection based on SHAP and fusion query strategies (WiFi, MeFi).","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":29,"slug":119},7,"Healthcare","healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]