Header logo is

Local adaptive subspace regression




Incremental learning of sensorimotor transformations in high dimensional spaces is one of the basic prerequisites for the success of autonomous robot devices as well as biological movement systems. So far, due to sparsity of data in high dimensional spaces, learning in such settings requires a significant amount of prior knowledge about the learning task, usually provided by a human expert. In this paper we suggest a partial revision of the view. Based on empirical studies, we observed that, despite being globally high dimensional and sparse, data distributions from physical movement systems are locally low dimensional and dense. Under this assumption, we derive a learning algorithm, Locally Adaptive Subspace Regression, that exploits this property by combining a dynamically growing local dimensionality reduction technique  as a preprocessing step with a nonparametric learning technique, locally weighted regression, that also learns the region of validity of the regression. The usefulness of the algorithm and the validity of its assumptions are illustrated for a synthetic data set, and for data of the inverse dynamics of human arm movements and an actual 7 degree-of-freedom anthropomorphic robot arm. 

Author(s): Vijayakumar, S. and Schaal, S.
Book Title: Neural Processing Letters
Volume: 7
Number (issue): 3
Pages: 139-149
Year: 1998

Department(s): Autonomous Motion
Bibtex Type: Article (article)

Cross Ref: p1031
Note: clmc
URL: http://www-clmc.usc.edu/publications/V/vijayakumar-NPL1998.pdf


  title = {Local adaptive subspace regression},
  author = {Vijayakumar, S. and Schaal, S.},
  booktitle = {Neural Processing Letters},
  volume = {7},
  number = {3},
  pages = {139-149},
  year = {1998},
  note = {clmc},
  crossref = {p1031},
  url = {http://www-clmc.usc.edu/publications/V/vijayakumar-NPL1998.pdf}