ICANN'07 - International Conference on Artificial Neural Networks

SPECIAL SESSIONS

Submission of papers to special sessions:
Papers submitted to Special Sessions must follow the same guidelines as the general conference papers. See the Paper Submission page for detailed requirements and instructions.

Important Dates:
March 23, 2007: Paper submission deadline (for all papers).


List of Special Sessions, Organizers and Main Topics

1. Complex-Valued Neural Networks

Igor Aizenberg, Texas A&M University-Texarkana, USA, igor.aizenberg(a.t)tamut.edu
Akira Hirose, University of Tokyo, Japan, ahirose(a.t)ee.t.u-tokyo.ac.jp
Jacek M. Zurada, University of Louisville, USA, jacek.zurada(a.t)louisville.edu

Papers that are related to all aspects of the CVNNs are invited. We seek contributions of theoretical but also of applied nature. We also welcome potential contributions form other areas that are on the boarders of the proposed scope.
Topics include, but are not limited to:

• Theoretical Aspects of CVNNs and Complex-Valued Activation Functions
• Complex-Valued Associative Memories
• Dynamics of Complex-Valued Neurons
• Learning Algorithms for CVNNs
• Feedforward CVNNs
• Spatiotemporal CVNNs Processing
• Frequency Domain CVNNs Processing
• Phase-Sensitive Signal Processing
• Chaos in Complex Domain
• Pattern Recognition, Classification and Time Series Prediction using CVCNNs
• Applications of CVNNs in Image Processing, Speech Processing, Robotics and Bioinformatics
• Quantum Computation and Quantum Neural Networks
• Quaternion and Clifford Networks


2. Emotion and Attention: Empirical Findings and Neural Models

Nienke Korsten, King’s College London, UK, nienke.korsten(a.t)kcl.ac.uk
John Taylor, King’s College London, UK, john.g.taylor(a.t)kcl.ac.uk

Papers related to the following topics are welcomed:

• Emotional functionality through neuromodulatory mechanisms
• Neural substructures underlying appraisal processes
• Attentional aspects of appraisal
• Empirical findings related to the mechanisms underlying human emotions / attention
• Neurally plausible models related to particular emotions (e.g. anger, fear)
• The interplay between attention and emotion
• Computational models of emotional influences on attentional processes
• …of attentional influences on emotional processes
• How to bridge the gap between appraisal and  emotions
• The involvement of emotions in decision making
• The learning of values of objects by dopamine-based processes
• The learning of values of affordances in terms of rewarded actions or rewarded objects
• The interaction between emotion and language


3. Meta-Learning

Norbert Jankowski, Nicolaus Copernicus University, Poland, norbert(a.t)is.umk.pl
Wlodzislaw Duch, Nicolaus Copernicus University, Poland, wduch(a.t)is.umk.pl
Krzysztof Grabczewski, Nicolaus Copernicus University, Poland, kgrabcze(a.t)is.umk.pl

Papers related to the following topics are welcomed:

• Criteria and methods for automated search for best data models.
• Extraction of knowledge guiding the meta-learning process, collection and re-use of information for learning.
• Strategies for search in parameter space in which data models are specified.
• Methods of discovering nontrivial solutions in reasonable time.
• Control and selection of optimal model complexity depending on various criteria.
• Meta-learning architectures: What is necessary to facilitate meta-search from the information exchange point of view? Methods and data structures of abstract information exchange in meta-learning. How to build, exploit and rearrange meta-knowledge?
• Intermediate testing to estimate models quality and to direct further learning.
• All meta-learning computational intelligence approaches are welcome, meta-learning is not limited to neural network models, although modular neural models are of great interest.

This session is NOT devoted to applications of already known algorithms or development of single algorithms.


4. Understanding and Creating Cognitive Systems

John Taylor, King’s College London, UK, john.g.taylor(a.t)kcl.ac.uk

The session will accept papers attempting to model human cognitive powers as well as those designing and testing cognitive systems based more loosely on such powers but attacking hard information processing tasks in novel ‘cognitive’ manners. More general papers will also be called for which consider cognition as composed of a broad raft of faculties: thinking, reasoning, imagining, planning, creating new concepts, perceiving and deciding on responses. At the highest level there is also the subtle phenomenon of consciousness, which cannot be left out of any definition of cognition (and which enters critically in various dictionary definitions of cognition). Thus implications of the development of specific cognitive systems for understanding consciousness will be welcomed.
Some of the basic questions needing to be studied in the papers presented at the session are:

• What is the nature of machine cognition or intelligence, as beyond pattern matching, classification and prediction?
• Is language essential to such machine intelligence and if so how is it to be implemented?
• How can we otherwise bridge the sub-symbolic - symbolic barrier? in this process
• Is embodiment crucial to such machine intelligence, and how complex should that be if it is regarded as crucial?
• What are the most appropriate industrial applications of such machine intelligence if and when it is attained?
• What are natural software architectures for creating such machine intelligence?
• What role should and would emotions play in such machine intelligence?
• What are the ethical problems involved in creating such machine intelligence?
• Is consciousness essential to the implementation of such machine intelligence?


5. Temporal synchronization and nonlinear dynamics in neural networks

Alessandro Villa, University of Grenoble, France, avilla(a.t)neuroheuristic.org

The session is aimed towards the understanding of nonlinear dynamics in neural circuits by theoretical, computational and experimental means. This concerns nonlinearities at synaptic, cellular and network levels. In particular, the session would like to emphasize the role of synchronization as carrier of information with respect to global activity patterns in the brain.
Papers related to the following topics are welcomed:

• Bifurcations in neural dynamics
• Computational neuroscience
• Temporal analysis of the spike train
• Detection of deterministic chaos in brain
• Topology of neural networks
• Synchronization of neural activity
• Oscillatory activity in cell assemblies
• Coupling of neural networks activity
• Spike timining dependent plasticity
• Spreading of activity waves in the central nervous system
• Synfire chains

Copyright © ICANN 2007
Instituto Superior de Engenharia do Porto