Highlights:<br>•\tTailBooster pairs anomaly detection with generation to synthesise valid extremes.<br>•\tA statistical layer isolates rare records to give the generator a tail signal.<br>•\tA learned operational envelope discards infeasible synthetic flight records.<br>•\tError on extreme air time and arrival delay prediction fell by up to 57 percent.<br>•\tThe fully data-driven design transfers to domains with no known governing rules.</p>\n","updatedAt":"2026-08-14T08:43:33.330Z","author":{"_id":"69e9c7add7407de81b825900","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/Hi92317VTdDpyeqaDPJCn.jpeg","fullname":"Karim Yehia Aly","name":"karimyehia92","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.8097794651985168},"editors":["karimyehia92"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/Hi92317VTdDpyeqaDPJCn.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.11951","authors":[{"_id":"6a7ed36b42823931a1f17803","name":"Karim Aly","hidden":false},{"_id":"6a7ed36b42823931a1f17804","name":"Alexei Sharpanskykh","hidden":false},{"_id":"6a7ed36b42823931a1f17805","name":"Jacco Hoekstra","hidden":false}],"publishedAt":"2026-08-12T00:00:00.000Z","submittedOnDailyAt":"2026-08-14T00:00:00.000Z","title":"TailBooster: A Dual-Layer Generative Framework for Extreme Value Augmentation with Operational Validity Enforcement","submittedOnDailyBy":{"_id":"69e9c7add7407de81b825900","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/Hi92317VTdDpyeqaDPJCn.jpeg","isPro":false,"fullname":"Karim Yehia Aly","user":"karimyehia92","type":"user","name":"karimyehia92"},"summary":"Extreme events in air transport, such as severe arrival delays and abnormal air times, cause cascading network disruptions with substantial operational, economic, and safety costs. Such events are rare in historical records, leaving insufficient training signal for machine learning models. Synthetic data augmentation offers a principled solution, but conventional generative models under-represent distributional tails and give no guarantee against operationally infeasible instances, such as a short air time paired with a long flight distance. No existing approach addresses both limitations for mixed-type tabular records. We propose TailBooster, a dual-layer generative framework combining generative modelling with two anomaly detection layers. A statistical layer extracts extremes via the interquartile range, supplying tail-concentrated training signal to dedicated generative models, here a Tabular Variational Autoencoder. A deep learning layer then applies autoencoder-based cleaning, discarding synthetic records that violate the operational envelope learned from historical data. The framework was evaluated on US flight records across five dimensions: diversity, statistical similarity, fidelity, operational validity, and utility, the latter two being the primary improvement targets. Data-driven cleaning markedly improved operational validity, while targeted augmentation enhanced utility for extreme-event prediction. Across six regression algorithms, training on the framework's records reduced Mean Absolute Error by 47-49% on extreme air time and 29-57% on extreme arrival delay prediction relative to conventional synthetic data, with comparable gains when real records were enriched with synthetic extremes. Being fully data-driven and model-agnostic, TailBooster extends to domains where extreme-event prediction is critical and domain-specific rules are unavailable.","upvotes":0,"discussionId":"6a7ed36c42823931a1f17806","projectPage":"https://www.sesarju.eu/projects/SynthAIR","ai_summary":"TailBooster uses a dual-layer generative framework with statistical tail extraction and deep autoencoder cleaning to synthesize operationally valid extreme air-transport events, substantially improving extreme-event prediction accuracy.","ai_keywords":["Tabular Variational Autoencoder","autoencoder-based cleaning","interquartile range","synthetic data augmentation","extreme-event prediction","mixed-type tabular data","operational validity"],"ai_summary_model":"thinkingmachines/Inkling-Small","organization":{"_id":"630e066483f64e3516754994","name":"TUDelft","fullname":"Technology University of Delft","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/61ac8f8a00d01045fca0ad2f/041dxbIduTXThGWs2AC3V.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"organization":{"_id":"630e066483f64e3516754994","name":"TUDelft","fullname":"Technology University of Delft","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/61ac8f8a00d01045fca0ad2f/041dxbIduTXThGWs2AC3V.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.11951.md","query":{}}">
TailBooster: A Dual-Layer Generative Framework for Extreme Value Augmentation with Operational Validity Enforcement
Abstract
TailBooster uses a dual-layer generative framework with statistical tail extraction and deep autoencoder cleaning to synthesize operationally valid extreme air-transport events, substantially improving extreme-event prediction accuracy.
Extreme events in air transport, such as severe arrival delays and abnormal air times, cause cascading network disruptions with substantial operational, economic, and safety costs. Such events are rare in historical records, leaving insufficient training signal for machine learning models. Synthetic data augmentation offers a principled solution, but conventional generative models under-represent distributional tails and give no guarantee against operationally infeasible instances, such as a short air time paired with a long flight distance. No existing approach addresses both limitations for mixed-type tabular records. We propose TailBooster, a dual-layer generative framework combining generative modelling with two anomaly detection layers. A statistical layer extracts extremes via the interquartile range, supplying tail-concentrated training signal to dedicated generative models, here a Tabular Variational Autoencoder. A deep learning layer then applies autoencoder-based cleaning, discarding synthetic records that violate the operational envelope learned from historical data. The framework was evaluated on US flight records across five dimensions: diversity, statistical similarity, fidelity, operational validity, and utility, the latter two being the primary improvement targets. Data-driven cleaning markedly improved operational validity, while targeted augmentation enhanced utility for extreme-event prediction. Across six regression algorithms, training on the framework's records reduced Mean Absolute Error by 47-49% on extreme air time and 29-57% on extreme arrival delay prediction relative to conventional synthetic data, with comparable gains when real records were enriched with synthetic extremes. Being fully data-driven and model-agnostic, TailBooster extends to domains where extreme-event prediction is critical and domain-specific rules are unavailable.
Community
Highlights:
• TailBooster pairs anomaly detection with generation to synthesise valid extremes.
• A statistical layer isolates rare records to give the generator a tail signal.
• A learned operational envelope discards infeasible synthetic flight records.
• Error on extreme air time and arrival delay prediction fell by up to 57 percent.
• The fully data-driven design transfers to domains with no known governing rules.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2608.11951 in a model README.md to link it from this page.
Cite arxiv.org/abs/2608.11951 in a dataset README.md to link it from this page.
Cite arxiv.org/abs/2608.11951 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.