AI & ML
Auto-differentiable data assimilation: Co-learning of states, dynamics, and filtering algorithms
arXiv stat.ML0 views
arXiv:2603.20891v2 Announce Type: replace Abstract: Data assimilation algorithms estimate the state of a dynamical system from partial observations, where the successful performance of these algorithms hinges on costly parameter tuning and on employing an accurate model for the dynamics. This paper introduces a framework for jointly learning the state, dynamics, and parameters of filtering algorithms in data assimilation through a process we refer to as auto-differentiable filtering. The framework leverages a theoretically motivated loss function that enables learning from partial, noisy observations via gradient-based optimization using auto-differentiation. We further demonstrate how several well-known data assimilation methods can be learned or tuned within this framework. To underscore the versatility of auto-differentiable filtering, we perform experiments on dynamical systems spanning multiple scientific domains, such as the Clohessy-Wiltshire equations from aerospace engineering, the Lorenz-96 system from atmospheric science, the generalized Lotka-Volterra equations from systems biology, and the two-layer quasi-geostrophic model from geophysical fluid dynamics. Finally, we provide guidelines for practitioners to customize our framework according to their observation model, accuracy requirements, and computational budget.
Source:https://arxiv.org/abs/2603.20891PreviousOn sparsity, extremal structure, and monotonicity properties of Wasserstein and Gromov-Wasserstein optimal transport plansNextmlr3torch: A Deep Learning Framework in R based on mlr3 and torch
Related posts

Delhi-based Bharat Semi, which is developing programmable image sensors combining capture, processing, and AI capabilities on a single chip, raised a $24M seed (Jagmeet Singh/TechCrunch)
0Techmeme

Russia is bringing young foreign influencers to occupied Ukraine, encouraging them to share their highly curated experiences, to legitimize its occupation (Linda Hourani/The Kyiv Independent)
0Techmeme

Startups unveil plans to develop floating offshore data centers to bypass land and power constraints, while experts warn of challenges like saltwater corrosion (Jackie Snow/Wall Street Journal)
0Techmeme

Analysis: global manufacturing, utility, and mining companies have shed $4.2T in market value since June as the AI "picks and shovels" trade swings into reverse (Clara Murray/Financial Times)
0Techmeme
Kernel prepatch 7.4-rc3
0LWN.net
Ask HN: Startups are spamming helpdesk chats now?
0Hacker News New