Learning to Recognize and Grasp Objects
Machine Learning - Special issue on learning in autonomous robots
Learning to Recognize and Grasp Objects
Autonomous Robots
Hierarchical Growing Cell Structures: TreeGCS
IEEE Transactions on Knowledge and Data Engineering
Classification of heart sounds using an artificial neural network
Pattern Recognition Letters
Two Modules of a Vision-Based Robotic System: Attention and Accumulation of Object Representations
RobVis '01 Proceedings of the International Workshop on Robot Vision
A self-organising network that grows when required
Neural Networks - New developments in self-organizing maps
The Evolving Tree—A Novel Self-Organizing Network for Data Analysis
Neural Processing Letters
Rule Extraction from Recurrent Neural Networks: A Taxonomy and Review
Neural Computation
Constructive Incremental Learning from Only Local Information
Neural Computation
Validating neural network-based online adaptive systems: a case study
Software Quality Control
Variations of the two-spiral task
Connection Science
A growing self-organizing network for reconstructing curves and surfaces
IJCNN'09 Proceedings of the 2009 international joint conference on Neural Networks
Magnification control in winner relaxing neural gas
Neurocomputing
Lyapunov analysis of neural network stability in an adaptive flight control system
SSS'03 Proceedings of the 6th international conference on Self-stabilizing systems
Extending the growing neural gas classifier for context recognition
EUROCAST'07 Proceedings of the 11th international conference on Computer aided systems theory
Learning-based robot vision: principles and applications
Learning-based robot vision: principles and applications
UAV swarm coordination using cooperative control for establishing a wireless communications backbone
Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems: volume 3 - Volume 3
Novelty detection for a neural network-based online adaptive system
COMPSAC-W'05 Proceedings of the 29th annual international conference on Computer software and applications conference
Analyzing large image databases with the evolving tree
ICAPR'05 Proceedings of the Third international conference on Advances in Pattern Recognition - Volume Part I
Learning to mimic motion of human arm and hand grabbing for constraint adaptation
DAGM'06 Proceedings of the 28th conference on Pattern Recognition
Learning deformations of human arm movement to adapt to environmental constraints
AMDO'06 Proceedings of the 4th international conference on Articulated Motion and Deformable Objects
Predicting with confidence – an improved dynamic cell structure
ICNC'05 Proceedings of the First international conference on Advances in Natural Computation - Volume Part I
An approach to v&v of embedded adaptive systems
FAABS'04 Proceedings of the Third international conference on Formal Approaches to Agent-Based Systems
Spike-timing-dependent construction
Neural Computation
Hi-index | 0.00 |
Dynamic cell structures (DCS) represent a family ofartificial neural architectures suited both for unsupervisedand supervised learning. They belong to the recently(Martinetz 1994) introduced class of topology representingnetworks (TRN) that build perfectly topology preservingfeature maps. DCS employ a modified Kohonen learningrule in conjunction with competitive Hebbian learning.The Kohonen type learning rule serves to adjust the synaptic weightvectors while Hebbian learning establishes a dynamic lateralconnection structure between the units reflecting the topologyof the feature manifold. In case of supervised learning, i.e.,function approximation, each neural unit implements a radialbasis function, and an additional layer of linear output unitsadjusts according to a delta-rule. DCS is the firstRBF-based approximation scheme attempting to concurrently learn andutilize a perfectly topology preserving map for improvedperformance. Simulations on a selection of CMU-Benchmarks indicatethat the DCS idea applied to the growing cell structurealgorithm (Fritzke 1993c) leads to an efficient and elegantalgorithm that can beat conventional models on similar tasks.