LAtent Phase Inference from Short time sequences using SHallow REcurrent Decoders (LAPIS-SHRED)

2026-04-01

AI Summary

This work addresses the challenge of reconstructing full spatiotemporal dynamics from extremely sparse observations confined to short time windows. The authors propose a three-stage modular framework that first pretrains a shallow recurrent decoder (SHRED) on simulation data to map sparse sensor time series into a structured latent space, then learns temporal propagation laws governing the latent states, and finally enables joint bidirectional reconstruction and prediction using only minimal, short-duration real-world observations. The approach supports data assimilation and multiscale reconstruction, accommodates extreme constraints such as single-frame terminal inputs, and features a lightweight, deployment-efficient architecture. Validation across six complex physical systems—including turbulence, combustion transients, multiscale propulsion, and satellite environmental fields—demonstrates its superior performance over existing methods.

Abstract

Reconstructing full spatio-temporal dynamics from sparse observations in both space and time remains a central challenge in complex systems, as measurements can be spatially incomplete and can be also limited to narrow temporal windows. Yet approximating the complete spatio-temporal trajectory is essential for mechanistic insight and understanding, model calibration, and operational decision-making. We introduce LAPIS-SHRED (LAtent Phase Inference from Short time sequence using SHallow REcurrent Decoders), a modular architecture that reconstructs and/or forecasts complete spatiotemporal dynamics from sparse sensor observations confined to short temporal windows. LAPIS-SHRED operates through a three-stage pipeline: (i) a SHRED model is pre-trained entirely on simulation data to map sensor time-histories into a structured latent space, (ii) a temporal sequence model, trained on simulation-derived latent trajectories, learns to propagate latent states forward or backward in time to span unobserved temporal regions from short observational time windows, and (iii) at deployment, only a short observation window of hyper-sparse sensor measurements from the true system is provided, from which the frozen SHRED model and the temporal model jointly reconstruct or forecast the complete spatiotemporal trajectory. The framework supports bidirectional inference, inherits data assimilation and multiscale reconstruction capabilities from its modular structure, and accommodates extreme observational constraints including single-frame terminal inputs. We evaluate LAPIS-SHRED on six experiments spanning complex spatio-temporal physics: turbulent flows, multiscale propulsion physics, volatile combustion transients, and satellite-derived environmental fields, highlighting a lightweight, modular architecture suited for operational settings where observation is constrained by physical or logistical limitations.

Problem

• spatio-temporal dynamics
• sparse observations
• short time sequences
• latent phase inference
• complex systems

Innovation

• latent phase inference
• shallow recurrent decoders
• spatio-temporal reconstruction
• hyper-sparse observations
• modular architecture