All talks will be held in lecture hall H6 at the western end of the main hall.
The time for a talk is 20 minutes plus 5 min discussion.
Monday, 2007-09-03 - from 16:15 to 17:30
Topographic Processing of Relational Data
|Presenting Author: Barbara Hammer|
|Authors: Barbara Hammer, Alexander Hasenfuß, Fabrice Rossi, Marc Strickert|
Recently, batch optimization schemes of the self-organizing map and neural gas have been modified to allow arbitrary distance measures.This principle is particularly suitable for complex applications where data are compared by means of problem-specific, possibly discrete metrics such as protein sequences. However, median variants do not allow a continuous update of prototype locations and their capacity is thus restricted. In this contribution, we consider the relational dual of batch optimization which can be formulated in terms of pairwise distances only such that an application to arbitrary distance matrices becomes possible. For SOM, a direct visualization of data is given by means of the underlying (euclidean or hyperbolic) lattice structure. For NG, pairwise distances of prototypes can be computed based on a given data matrix only, such that subsequent mapping by means of multidimensional scaling can be applied.
Visual mining in music collections with Emergent SOM
|Presenting Author: Sebastian Risi|
|Authors: Sebastian Risi, Fabian Mörchen, Alfred Ultsch, Pascal Lehwark|
Different methods of organizing large collections of music with databionic mining techniques are described. The Emergent Self-Organizing Map is used to cluster and visualize similar artists
and songs. The first method is the MusicMiner system that utilizes semantic
descriptions learned from low level audio features for each song. The second
method uses tags that have been assigned to music artists by the users of the
social music platform Last.fm. For both methods we demonstrate the
visualization capabilities of the U-Map. An intuitive browsing of large music collections is offered based on the paradigm of topographic maps. The semantic concepts behind the features enhance the interpretability of the maps.
SOM-based experience representation for Dextrous Grasping
|Presenting Author: Jan Frederik Steffen|
|Authors: Jan Frederik Steffen, Robert Haschke, Helge Ritter|
We present an approach to dextrous robot grasping which combines a purely
tactile-driven algorithm with an implicit representation of grasp
experience to yield an algorithm which can handle arbitrary, partially
unknown grasp situations. During the grasp movement, the obtained contact
information is used to dynamically
adapt the grasping control by targeting the best matching posture from the
experience base. Thus, the robot recalls and actuates a grasp it
already successfully performed in a similar tactile context. To efficiently
represent the experience, we introduce the Grasp Manifold assuming
that grasp postures form a smooth manifold in hand posture space. We
present a simple way of providing approximations of Grasp Manifolds
using Self-Organising Maps (SOMs) and study the properties of the
represented grasp manifolds concerning their smoothness and robustness
against clustered training data.