Learning Nonlinear Dynamic Models

2009/05/20 by Langford, John, Salakhutdinov, Ruslan, Zhang, Tong
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.0905.3369

Abstract

We present a novel approach for learning nonlinear dynamic models, which leads to a new set of tools capable of solving problems that are otherwise difficult. We provide theory showing this new approach is consistent for models with long range structure, and apply the approach to motion capture and high-dimensional video data, yielding results superior to standard alternatives.

Related

• Uncovering Closed-form Governing Equations of Nonlinear Dynamics from Videos
• Machine Learning in Nonlinear Dynamical Systems
• Modeling Nonlinear Dynamics from Videos
• Learning Physics From Video: Unsupervised Physical Parameter Estimation for Continuous Dynamical Systems
• Learning Reduced Nonlinear State-Space Models: an Output-Error Based Canonical Approach