learning linear dynamical systems
Summary
We can perform maximum likelihood estimation for the parameters of a linear dynamical system using the EM algorithm. The E step involves running a Kalman smoother, and the M step involves maximum likelihood inference in multivariate Gaussians.
Context
This concept has the prerequisites:
- linear dynamical systems
- Expectation-Maximization algorithm (We use the EM algorithm to learn the parameters.)
- Kalman smoother (the E step requires computing marginals using a Kalman smoother.)
- maximum likelihood: multivariate Gaussians (The M step involves maximum likelihood estimation of multivariate Gaussians.)
Core resources (we're sorry, we haven't finished tracking down resources for this concept yet)
Supplemental resources (the following are optional, but you may find them useful)
-Paid-
→ Pattern Recognition and Machine Learning
A textbook for a graduate machine learning course, with a focus on Bayesian methods.
Location:
Section 13.3.2, pages 642-644
See also
-No Additional Notes-