Rapid learning with parametrized self-organizing maps

Walter, Jörg A.; Ritter, Helge

doi:10.1016/0925-2312(95)00117-4

Cited by 79 publications

(36 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…[1]). One more recent method, Unsupervised Kernel Regression (UKR, [6,4]), can be seen as a successor bridging between earlier "Parametrised SOM" (PSOM, [11]) and kernel methods (e.g. [8]).…”

Section: Introductionmentioning

confidence: 99%

Towards Semi-supervised Manifold Learning: UKR with Structural Hints

Steffen

Klanke

Vijayakumar

et al. 2009

Advances in Self-Organizing Maps

Self Cite

View full text Add to dashboard Cite

Abstract. We explore generic mechanisms to introduce structural hints into the method of Unsupervised Kernel Regression (UKR) in order to learn representations of data sequences in a semi-supervised way. These new extensions are targeted at representing a dextrous manipulation task. We thus evaluate the effectiveness of the proposed mechanisms on appropriate toy data that mimic the characteristics of the aimed manipulation task and thereby provide means for a systematic evaluation.

show abstract

Section: Introductionmentioning

confidence: 99%

Towards Semi-supervised Manifold Learning: UKR with Structural Hints

Steffen

Klanke

Vijayakumar

et al. 2009

Advances in Self-Organizing Maps

Self Cite

View full text Add to dashboard Cite

show abstract

“…The network could be considered as forming a "dynamic map" of the arm, because it is able to represent all geometrically possible configurations. However, it requires a very small number of units compared to the usual topological, e.g., Kohonen type mappings (e.g., Walter & Ritter, 1996).…”

Section: Cognitive Systemsmentioning

confidence: 99%

The evolution of cognition—a hypothesis

Cruse¹

2003

Cognitive Science

View full text Add to dashboard Cite

Behavior may be controlled by reactive systems. In a reactive system the motor output is exclusively driven by actual sensory input. An alternative solution to control behavior is given by "cognitive" systems capable of planning ahead. To this end the system has to be equipped with some kind of internal world model. A sensible basis of an internal world model might be a model of the system's own body. I show that a reactive system with the ability to control a body of complex geometry requires only a slight reorganization to form a cognitive system. This implies that the assumption that the evolution of cognitive properties requires the introduction of new, additional modules, namely internal world models, is not justified. Rather, these modules may already have existed before the system obtained cognitive properties. Furthermore, I discuss whether the occurrence of such world models may lead to systems having internal perspective.

show abstract

“…The adaptive subspace SOM (ASSOM) has been proposed to combine principal component learning and the SOM to map data with reduced feature space, in order to form translation-, rotation-and scale-invariant filters [51,52]. The parameterized SOM (PSOM) has been proposed to extend SOM for continuous mapping using basis functions on the grid to interpolate the map [112]. The stochastic SOM [26] defines a topographic mapping from a Bayesian framework and a Markov chain encoder, and further explains the stochastic annealing effect in SOM.…”

Section: Extensions and Links With Other Learning Paradigmsmentioning

confidence: 99%