Complex, temporally evolving phenomena, from climate to brain activity, are governed by dynamical systems (DS). DS reconstruction (DSR) seeks to infer generative surrogate models of these from observed data, reproducing their long-term behavior. Existing DSR approaches require purpose-training for any new system observed, lacking the zero-shot and in-context inference capabilities known from LLMs. Here we introduce DynaMix, a novel multivariate ALRNN-based mixture-of-experts architecture pre-trained for DSR, the first DSR model able to generalize zero-shot to out-of-domain DS. Just from a provided context signal, without any re-training, DynaMix faithfully forecasts the long-term evolution of novel DS where existing time series (TS) foundation models, like Chronos, fail -- at a fraction of the number of parameters (0.1%) and orders of magnitude faster inference times. DynaMix outperforms TS foundation models in terms of long-term statistics, and often also short-term forecasts, even on real-world time series, like traffic or weather data, typically used for training and evaluating TS models, but not at all part of DynaMix' training corpus. We illustrate some of the failure modes of TS models for DSR problems, and conclude that models built on DS principles may bear a huge potential also for advancing the TS prediction field.

翻译：复杂的时间演化现象，从气候到大脑活动，都受动态系统（DS）支配。动态系统重构（DSR）旨在从观测数据中推断这些系统的生成式代理模型，以复现其长期行为。现有的DSR方法需要对任何新观测系统进行针对性训练，缺乏类似大语言模型（LLM）所具备的零样本与上下文推断能力。本文提出DynaMix，一种基于多元ALRNN的专家混合架构，专为DSR进行预训练，成为首个能够零样本泛化至域外动态系统的DSR模型。仅需提供上下文信号且无需重新训练，DynaMix即可准确预测新动态系统的长期演化——现有时间序列（TS）基础模型（如Chronos）在此任务上失效，而DynaMix仅需极少参数量（0.1%）且推理速度提升数个数量级。在长期统计特性方面，DynaMix优于TS基础模型，甚至在短期预测方面也常表现更佳，即使在交通或气象等真实世界时间序列数据上（这些数据通常用于训练和评估TS模型，但完全未包含在DynaMix的训练语料中）也是如此。我们展示了TS模型在DSR问题中的若干失效模式，并指出基于动态系统原理构建的模型可能对推进时间序列预测领域发展具有巨大潜力。