Nonlinear Latent Variable Models for Video Sequences
Author(s)
rahimi, ali
recht, ben
darrell, trevor
Date Issued
June 6, 2005
Series/Report no.
Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory
Abstract
Many high-dimensional time-varying signals can be modeled as a sequence of noisy nonlinear observations of a low-dimensional dynamical process. Given high-dimensional observations and a distribution describing the dynamical process, we present a computationally inexpensive approximate algorithm for estimating the inverse of this mapping. Once this mapping is learned, we can invert it to construct a generative model for the signals. Our algorithm can be thought of as learning a manifold of images by taking into account the dynamics underlying the low-dimensional representation of these images. It also serves as a nonlinear system identification procedure that estimates the inverse of the observation function in nonlinear dynamic system. Our algorithm reduces to a generalized eigenvalue problem, so it does not suffer from the computational or local minimum issues traditionally associated with nonlinear system identification, allowing us to apply it to the problem of learning generative models for video sequences.
Subjects
AI
Manifold learning,nonlinear system identification
unsupervised learning
Persistent DSpace Link
