English

Mixed LICORS: A Nonparametric Algorithm for Predictive State Reconstruction

Methodology 2015-07-27 v2 Machine Learning

Abstract

We introduce 'mixed LICORS', an algorithm for learning nonlinear, high-dimensional dynamics from spatio-temporal data, suitable for both prediction and simulation. Mixed LICORS extends the recent LICORS algorithm (Goerg and Shalizi, 2012) from hard clustering of predictive distributions to a non-parametric, EM-like soft clustering. This retains the asymptotic predictive optimality of LICORS, but, as we show in simulations, greatly improves out-of-sample forecasts with limited data. The new method is implemented in the publicly-available R package "LICORS" (http://cran.r-project.org/web/packages/LICORS/).

Keywords

Cite

@article{arxiv.1211.3760,
  title  = {Mixed LICORS: A Nonparametric Algorithm for Predictive State Reconstruction},
  author = {Georg M. Goerg and Cosma Rohilla Shalizi},
  journal= {arXiv preprint arXiv:1211.3760},
  year   = {2015}
}

Comments

11 pages; AISTATS 2013

R2 v1 2026-06-21T22:39:18.197Z