Learning Nonlinear Dynamic Models

TL;DR

This paper introduces a new method for learning nonlinear dynamic models, demonstrating theoretical consistency and superior performance on motion capture and high-dimensional video data compared to existing methods.

Contribution

The paper presents a novel approach for learning nonlinear dynamic models with proven consistency and improved results on complex real-world data.

Findings

The approach is consistent for models with long-range structure.

It outperforms standard alternatives on motion capture data.

It achieves superior results on high-dimensional video data.

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.