J-STAGE Home  >  Publications - Top  > Bibliographic Information

Transactions of the Japanese Society for Artificial Intelligence
Vol. 26 (2011) No. 6 P 638-648

Language:

http://doi.org/10.1527/tjsai.26.638

Original Paper

In this paper, we present an algorithm for learning non-linear dynamical systems which works by aligning local linear models, based on a probabilistic formulation of subspace identification. This is achieved by the fusion of the recent works in the fields of machine learning and system control. Because the procedure for constructing a state sequence in subspace identification can be interpreted as the Canonical Correlation Analysis(CCA) between past and future observation sequences, we can derive a latent variable representation for this problem. Therefore, as in a similar manner to the recent works on learning a mixture of probabilistic models, we obtain a framework for constructing a state space by aligning local linear coordinates. This leads to a prominent algorithm for learning non-linear dynamical systems. Finally, we apply our method to motion capture data and telemetry data, and then show how our algorithm works well.

Copyright © 2011 JSAI (The Japanese Society for Artificial Intelligence)

Article Tools

Share this Article