Breakthrough mathematics for dynamical systems and data

这一奖学金的目的是在我们用来从动力系统和动力数据中学习可操作信息的数学上产生一种阶段性的变化。该项目使用开创性的算子理论方法来分析高维系统和时空数据，期望在复杂系统的建模方面产生新的知识，并为无监督的机器学习提供新的途径。这一奖学金的预期成果包括一批新的数学和实用的下一代算法，以发现复杂动态系统和数据中隐藏的人类可理解的模式。这将为科学的许多领域提供重大的普遍利益，包括阐明气候和社会动态的基础机制。

This fellowship aims to create a step change in the mathematics we use to learn actionable information from dynamical systems and dynamical data. Using a groundbreaking, operator-theoretic approach to analyse high dimensional systems and spatiotemporal data, this project expects to generate new knowledge in the modelling of complex systems and new pathways for unsupervised machine learning. Expected outcomes of this fellowship include a tranche of new mathematics and practical next-generation algorithms to discover hidden human-understandable patterns in complex dynamical systems and data. This should provide significant universal benefits to many areas of science, including elucidating mechanisms underlying climate and social dynamics.