New Challenges in Neural Computation (NC2)
[Home][Call for Papers][Program]
Neural computation and biologically inspired data processing systems constitute essential topics in
artificial intelligence accompanied by a well established theoretical foundation and numerous successful
applications in science and industry. Although some of the most popular methods such as
learning for multilayer perceptrons, associative memories, or
self-organizing maps are well established and readily available in commercial tools,
modern information processing continues to pose challenging tasks to the field which are far from being
solved. Not only the amount of data explodes in virtually all application areas but
also their complexity with regard to dimensionality, structural variety, and multimodality,
such that models have
to deal with very large and heterogeneous data sets. At the same time, the tasks
become more and more complex, moving from simple classification or prediction in pattern recognition
to involved learning scenarios in dynamic environments with no explicit single objective,
such that models can no longer be based on simple error measures.
Humans are capable of handling complex situations and tasks by means of a combination
of different paradigms, whereas existing neural systems mostly mirror only one or a few facets
of the whole game.
The goal of the workshop is to figure out paradigms, concepts, and models to extend
neural systems to complex situations and tasks and to identify good
benchmark scenarios in which to test advanced capacities of model systems.
Concrete problems tackled in the workshop include the following:
neural models for very large data sets, complex learning scenarios and data structures,
principled mathematical models which address scenarios beyond classical classification, regression, clustering, etc.,
principled cognitive paradigms which help to design complex neural systems,
suitable application scenarios and evaluation procedures to benchmark
the capability of complex neural systems.
Submissions should be prepared as pdf files and should not
exceed 8 pages. Final manuscripts should be according to the
LNCS style as specified in the
instructions on the
KI call for papers.
Papers should be send to
Barbara Hammer by
e-mail (email@example.com) until
Abstracts of the papers will be available at the KI-workshop.
Further, accepted submissions will be published as
Machine Learning Report.
For further questions, please contact the workshop organizers
- Barbara Hammer, CITEC, Faculty of Technology, Bielefeld University, D-33594 Bielefeld, firstname.lastname@example.org
- Thomas Villmann, Mathematics, University of Applied Sciences Mittweida,