Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

12

ICML
2010
IEEE

favoriteEmaildiscussreport

202views Machine Learning» more ICML 2010»

Learning the Linear Dynamical System with ASOS

13 years 5 months ago

Learning the Linear Dynamical System with ASOS

Download www.cs.toronto.edu

We develop a new algorithm, based on EM, for learning the Linear Dynamical System model. Called the method of Approximated Second-Order Statistics (ASOS) our approach achieves dramatically superior computational performance over standard EM through its use of approximations, which we justify with both intuitive explanations and rigorous convergence results. In particular, after an inexpensive precomputation phase, the iterations of ASOS can be performed in time independent of the length of the training dataset.

James Martens

Real-time Traffic

ICML 2010 | Inexpensive Precomputation Phase | Linear Dynamical System | Machine Learning | Superior Computational Performance |

claim paper

Related Content

» Learning NonLinear Dynamical Systems by Alignment of Local Linear Models

» Learning linear dynamical systems without sequence information

» Nonparametric Bayesian Learning of Switching Linear Dynamical Systems

» Bayesian Nonparametric Inference of Switching Dynamic Linear Models

» A Constraint Generation Approach to Learning Stable Linear Dynamical Systems

» Cutandstitch efficient parallel learning of linear dynamical systems on smps

» Modeling Sensorimotor Learning with Linear Dynamical Systems

» Sparse Coding of Linear Dynamical Systems with an Application to Dynamic Texture Recogniti...

» Kernel Predictive Linear Gaussian models for nonlinear stochastic dynamical systems

Post Info
More Details (n/a)

Added	09 Nov 2010
Updated	09 Nov 2010
Type	Conference
Year	2010
Where	ICML
Authors	James Martens

Comments (0)