Beyond the Norm: A Survey of Synthetic Data Generation for Rare Events

2025-06-04

AI Summary

Extreme events—such as stock market crashes, earthquakes, and pandemics—are rare, catastrophic, and exhibit system-wide propagation, leading to severe data scarcity that undermines data-driven modeling. To address this, we present the first systematic survey of synthetic data generation methods tailored to extreme events and propose the first dedicated generative framework for extremely rare events. We design a customized evaluation suite encompassing statistical fidelity, dependency preservation, visual plausibility, and task-oriented utility, rigorously analyzing metric validity under heavy-tailed distributions. Our framework unifies generative models (GANs, diffusion models, VAEs), large language models, statistical modeling, and targeted resampling strategies. We curate benchmark datasets across finance, meteorology, geoscience, and epidemiology, identifying underexplored domains—including behavioral finance, wildfire dynamics, and windstorm modeling—and distill key open challenges to advance the reliability and practicality of extreme-event modeling.

Technology Category

• Natural Language Processing: Generation
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Web Mining and Content Analysis: Web data generation and simulation
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Extreme events, such as market crashes, natural disasters, and pandemics, are rare but catastrophic, often triggering cascading failures across interconnected systems. Accurate prediction and early warning can help minimize losses and improve preparedness. While data-driven methods offer powerful capabilities for extreme event modeling, they require abundant training data, yet extreme event data is inherently scarce, creating a fundamental challenge. Synthetic data generation has emerged as a powerful solution. However, existing surveys focus on general data with privacy preservation emphasis, rather than extreme events' unique performance requirements. This survey provides the first overview of synthetic data generation for extreme events. We systematically review generative modeling techniques and large language models, particularly those enhanced by statistical theory as well as specialized training and sampling mechanisms to capture heavy-tailed distributions. We summarize benchmark datasets and introduce a tailored evaluation framework covering statistical, dependence, visual, and task-oriented metrics. A central contribution is our in-depth analysis of each metric's applicability in extremeness and domain-specific adaptations, providing actionable guidance for model evaluation in extreme settings. We categorize key application domains and identify underexplored areas like behavioral finance, wildfires, earthquakes, windstorms, and infectious outbreaks. Finally, we outline open challenges, providing a structured foundation for advancing synthetic rare-event research.

Problem

• Synthetic data generation for rare extreme events
• Addressing scarcity of training data for extreme events
• Evaluating models for heavy-tailed distributions in extreme events

Innovation

• Generative modeling for heavy-tailed distributions
• LLMs enhanced by statistical theory
• Tailored evaluation framework for extreme events