Back
Locally weighted learning for control
Lazy learning methods provide useful representations and training algorithms for learning about complex phenomena during autonomous adaptive control of complex systems. This paper surveys ways in which locally weighted learning, a type of lazy learning, has been applied by us to control tasks. We explain various forms that control tasks can take, and how this affects the choice of learning paradigm. The discussion section explores the interesting impact that explicitly remembering all previous experiences has on the problem of learning to control. Keywords: locally weighted regression, LOESS, LWR, lazy learning, memory-based learning, least commitment learning, forward models, inverse models, linear quadratic regulation (LQR), shifting setpoint algorithm, dynamic programming.
@article{Atkeson_AIR_1998, title = {Locally weighted learning for control}, booktitle = {Artificial Intelligence Review}, abstract = {Lazy learning methods provide useful representations and training algorithms for learning about complex phenomena during autonomous adaptive control of complex systems. This paper surveys ways in which locally weighted learning, a type of lazy learning, has been applied by us to control tasks. We explain various forms that control tasks can take, and how this affects the choice of learning paradigm. The discussion section explores the interesting impact that explicitly remembering all previous experiences has on the problem of learning to control. Keywords: locally weighted regression, LOESS, LWR, lazy learning, memory-based learning, least commitment learning, forward models, inverse models, linear quadratic regulation (LQR), shifting setpoint algorithm, dynamic programming.}, volume = {11}, number = {1-5}, pages = {75-113}, year = {1997}, note = {clmc}, slug = {atkeson_air_1998}, author = {Atkeson, C. G. and Moore, A. W. and Schaal, S.}, crossref = {p43}, url = {http://www-clmc.usc.edu/publications/A/atkeson-AIR-II-1997.pdf} }