Overview

Latent dynamics modeling studies how to represent a changing system with hidden states and learn how those states evolve from observations such as video, sensor streams, or sparse measurements. Common approaches include autoencoders with state-space models, neural ODEs and SDEs for continuous-time or stochastic change, Koopman operators for approximately linear evolution, and JEPA-style world models for action-conditioned prediction. These models support forecasting and control in robotics and physical simulation, as well as analysis of biological, clinical, and neural time series. A central challenge is that good one-step prediction or reconstruction does not guarantee reliable long-horizon rollouts, useful control, or recovery of meaningful physical mechanisms; work therefore explores stability, identifiability, and physics-based structure while balancing these constraints against model flexibility.

Latest papers

CardsList

274 papers27 papers in the last four weeks, down 50% on the four weeks before. 0.2% of all new papers.

  1. Latent Evolution Operator (LEO): State-to-State Forecasting of Parametric PDEs

    May 14, 2025Khalid Rafiq, Wenjing Liao, Aditya G. NairPDE Surrogate ModelingPDE Operator Learning

  2. Stochastic Optimal Control for Continuous-Time fMRI Representation Learning

    Feb 7, 2025Joonhyeong Park, Byoungwoo Park, Chang-Bae Bang +4Latent Dynamics ModelingSelf-Supervised Time Series Representation Learning

  3. High-Dimensional Markov-Switching Ordinary Differential Processes

    Dec 30, 2024Katherine Tsai, Mladen Kolar, Sanmi KoyejoLatent Dynamics ModelingDynamic Graph Learning

  4. Automated Discovery of Operable Dynamics from Videos

    Oct 14, 2024Kuang Huang, Dong Heon Cho, Boyuan ChenDynamical SystemsLatent Dynamics Modeling