Interpretable Cross-Sphere Multiscale Deep Learning Predicts ENSO Skilfully Beyond 2 Years
El Ni\~no-Southern Oscillation (ENSO) exerts global climate and societal impacts, but real-time prediction with lead times beyond one year remains challenging. Dynamical models suffer from large biases and uncertainties, while deep learning struggles with interpretability and multi-scale dynamics. Here, we introduce PTSTnet, an interpretable model that unifies dynamical processes and cross-scale spatiotemporal learning in an innovative neural-network framework with physics-encoding learning. PTSTnet produces interpretable predictions significantly outperforming state-of-the-art benchmarks with lead times beyond 24 months, providing physical insights into error propagation in ocean-atmosphere interactions. PTSTnet learns feature representations with physical consistency from sparse data to tackle inherent multi-scale and multi-physics challenges underlying ocean-atmosphere processes, thereby inherently enhancing long-term prediction skill. Our successful realizations mark substantial steps forward in interpretable insights into innovative neural ocean modelling.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Extracting the Multiscale Causal Backbone of Brain Dynamics
The bulk of the research effort on brain connectivity revolves around statistical associations among brain regions, which do not directly relate to the causal mechanisms governing brain dynamics. Here we propose the mult…
Functional ConnectivityCurvature-Conditioned Multiscale Momentum with Sphere Constraints for LLM Pretraining
Pretraining accounts for a large fraction of the total computational cost in LLM training. However, noise-dominant gradients and the highly ill-conditioned loss landscape bring severe challenges. Although modern adaptive…
Multiscale cortical morphometry reveals pronounced regional and scale-dependent variations across the lifespan
Motivation: Characterising the changes in cortical morphology across the lifespan is fundamental for a range of research and clinical applications. Most studies to date have found a monotonic decrease in commonly used mo…
Interpretable learning of effective dynamics for multiscale systems
The modeling and simulation of high-dimensional multiscale systems is a critical challenge across all areas of science and engineering. It is broadly believed that even with today's computer advances resolving all spatio…
GenSP: Consistent Spherical Parameterization via Learning Shape Generative Models
We introduce GenSP, a data-driven framework that learns consistent spherical parameterizations across a collection of genus-0 shapes. Instead of optimizing the parameterization of each shape independently, our method lea…