用上下文学习预测系统突变，无需重新训练即可泛化
In-context learning to predict critical transitions in dynamical systems
- 基于先验数据拟合的网络，利用不同规模上下文进行突变预测
- 在模拟、真实场景中均实现顶尖早期预警效果，支持零样本推理
- 适合需要快速适应新系统的复杂动力学研究者
临界转变——系统动态中突然且常不可逆的变化——广泛存在于人类与自然系统中，常带来灾难性后果。现实世界中此类转变的观测数据极为稀缺，制约了可靠早期预警系统的发展。传统统计与谱分析指标（如方差上升）在数据有限、噪声相关等现实条件下易失效；现有深度学习分类器亦无法超越训练数据分布的外推能力。本文提出TipPFN，一种基于上下文学习（ICL）的框架，通过在新型合成数据生成器上训练（该生成器基于典型分岔场景并耦合多样化随机动力学），可灵活利用不同规模、复杂度与维度的上下文信息，推断系统接近临界转变的程度。实验表明，该方法在未见过的突变情景、仿真到现实迁移以及真实观测中，均实现了稳健的、领先水平的早期检测，在ICL与零样本设置下表现优异。
原文摘要 · Abstract (English)
Critical transitions - abrupt, often irreversible changes in system dynamics - arise across human and natural systems, often with catastrophic consequences. Real-world observations of such shifts remain scarce, preventing the development of reliable early warning systems. Conventional statistical and spectral indicators, such as increasing variance, tend to fail under realistic conditions of limited data and correlated noise, whereas existing deep learning classifiers do not extrapolate beyond their training data distribution. In this work, we introduce TipPFN, an in-context learning (ICL) framework that uses a prior-data fitted network to infer a system's proximity to a critical transition. Trained on our novel synthetic data generator, which is based on canonical bifurcation scenarios coupled to diverse, randomized stochastic dynamics, TipPFN flexibly capitalizes on contexts of various sizes, complexity and dimensionalities. We demonstrate robust, state-of-the-art early detection of critical transitions in previously unseen tipping regimes, sim-to-real examples, and real-world observations in both ICL and zero-shot settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。