Abstract
We describe a hierarchical scheme for rapid adaptation of context dependent “skills”. The underlying idea is to first invest some learning effort to specialize the learning system to become a rapid learner for a restricted range of contexts. This is achieved by constructing a “Meta-mapping” that replaces an slow and iterative context adaptation by a “one-shot adaptation”, which is a context-dependent skill-reparameterization.
The notion of “skill” is very general and includes a task specific, hand-crafted function mapping with context dependent parameterization, a complex control system, as well as a general learning system.
A representation of a skill that is particularly convenient for the investment learning approach is by a Parameterized Self-Organizing Map (PSOM). Its direct constructability from even small data sets significantly simplifies the investment learning stage; its ability to operate as a continuous associative memory allows to represent skills in the form of “multi-way” mappings (relations) and provides an automatic mechanism for sensor data fusion.
We demonstrate the concept in the context of a (synthetic) vision task that involves the associative completion of a set of feature locations and the task of one-shot adaptation of the transformation between world and object coordinates to a changed camera view of the object.
Preview
Unable to display preview. Download preview PDF.
References
B. Brunner, K. Arbter, and G. Hirzinger. Task directed programming of sensor based robots. In Intelligent Robots and Systems (IROS-94), pages 1081–1087, September 1994.
K. Fu, R. Gonzalez, and C. Lee. Robotics: Control, Sensing, Vision, and Intelligence. McGraw-Hill, 1987.
K. Hornik, M. Stinchcombe, and H. White. Multilayer feedforward networks are universal approximators. Neural Networks, 2:359–366, 1989.
M. I. Jordan and R. A. Jacobs. Hierarchical mixtures of experts and the EM algorithm. Neural Computation, 6(2):181–214, 1994.
T. Kohonen. Self-Organizing Maps, volume 30 of Springer Series in Information Sciences. Springer, Berlin, Heidelberg, 1995.
H. Ritter. Parametrized self-organizing maps. In S. Gielen and B. Kappen, editors, Proc. Int. Conf. on Artificial Neural Networks (ICANN-93), Amsterdam, pages 568–575. Springer Verlag, Berlin, 1993.
J. Walter and H. Ritter. Investment learning with hierarchical PSOM. In NIPS *95, page (in press). MIT Press, 1995.
J. Walter and H. Ritter. Local PSOMs and Chebyshev PSOMs — improving the parametrised self-organizing maps. In Proc. Int. Conf. on Artificial Neural Networks (ICANN-95), Paris, volume 1, pages 95–102, 1995.
J. Walter and H. Ritter. Rapid learning with parametrized self-organizing maps. Neuro-computing, Special Issue, (in press), 1996.
J. A. Walter. Rapid Learning in Robotics. Infix Verlag, Sankt Augustin, Germany, 1996. (in preparation).
H. White. Learning in artificial neural networks: a statistical perspective. Neural Computation, 1:425–464, 1989.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Walter, J., Ritter, H. (1996). Associative completion and investment learning using PSOMs. In: von der Malsburg, C., von Seelen, W., Vorbrüggen, J.C., Sendhoff, B. (eds) Artificial Neural Networks — ICANN 96. ICANN 1996. Lecture Notes in Computer Science, vol 1112. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-61510-5_30
Download citation
DOI: https://doi.org/10.1007/3-540-61510-5_30
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61510-1
Online ISBN: 978-3-540-68684-2
eBook Packages: Springer Book Archive